fibrin and fibrinogen degradation products in plasma
TRANSCRIPT
FIBRIN AND FIBRINOGEN DEGRADATION
PRODUCTS IN PLASMA
Clinical and methodological studies using enzyme immuno assays
CIP - GEGEVENS KONINKL!JKE BIBLIOTiffiEK, DEN HAAG
Kroneman, Herman
Fibrin and fibrinogen degradation products in plasma : clinical and methodological studies using enzyme immuno assays I Herman Kroneman.- [ S.l. : s.n. ]. -Ill. Thesis Rotterdam. - With summary in Dutch. ISBN 90-9004926-6 Subject headings: bloodplasma I bloodcoagulationfacrors.
© H. Kroneman, !992
All rights reserved. No pan of this book may be reproduced or used in any way without the prior written permission from the author.
FIBRIN AND FIBRINOGEN DEGRADATION
PRODUCTS IN PLASMA
Clinical and methodological studies using enzyme immune assays
FIBRINE EN FIBR!NOGEEN AFBRAAKPRODUKTEN IN PLASMA
Klinische en methodologische studies met enzyme immune assays
PROEFSCHRIFT
TER VERKRUG!NG VAN DE GRAAD VAN DOCTOR AAN DE ERASMUS UN!VERSITE!T VAN ROTTERDAM
OP GEZAG VAN DE RECTOR MAGN!F!CUS PROF. DR. C.J. RIJNVOS
EN VOLGENS BESLU!T VAN HET COLLEGE VAN DEKANEN. DE OPENBARE VERDEDIGING ZAL PLAATSVINDEN OP
WOENSDAG 8 APRIL 1992 OM 15.45 UUR
door
HERMAN KRONEMAN
geboren te Harderwijk
PROMOTIECOMMISSIE
PROMOTOR:
CO-PROMOTOR:
OVERIGE LEDEN:
PROF. J.H.P. WILSON
DR. E.A.R. KNOT
PROF.DR.IR. J.D.F. HABBEMA PROF.DR. H. VAN URK PROF.DR. B. LOWENBERG
This study was performed at the department of Internal Medicine II of the University Hospital Rotterdam-Dijkzigt and the Center for Clinical Decision Analysis of the Erasmus University, Rotterdam. Financial support during the studies and for the publication of this thesis was kindly provided by the Camelis Visser Foundation, Organon Teknika, Kabi Phannacia Nederland, Glaxo, Boehringer Mannheim and Mere, Sharp & Dohme.
Financial support by the Netherlands Heart Foundation for the publication of this thesis is greatfully acknowlegded.
~ Gedrukt door: Drukkerij Haveka B.V., Alblasserdam
CONTENTS
Introduction.
Chapter I.
Monoclonal antibody-based plasma assays for fibrin(ogen)
and derivatives, and their clinical relevance.
Chapter 2.
Correlations between plasma levels of fibrin(ogen)
derivatives as quantified by different assays
based on monoclonal antibodies.
Chapter 3.
Diagnostic value of D-dimer for deep venous
thrombosis in outpatients.
Chapter 4.
Fibrin and fibrinogen degradation products in plasma
during gram-negative septic shock.
Chapter 5.
Pharmacokinetics of low molecular weight heparin and
unfractionated heparin during elective aortobifemoral
bypass grafting.
Chapter 6.
Diagnostic value of quantitative tests for fibrin
degradation products in deep venous thrombosis.
Chapter 7.
Discussion
Summary
Samenvatting
Acknowledgements
Curriculum vitae
13
65
81
93
105
125
145 147
151
155 159
INTRODUCTION
INTRODUCTION
The haemostatic mechanism is a dynamic balance which can basically be described as
the equilibrium between fibrin formation (coagulation) and fibrin lysis (fibrinolysis).
A schematic presentation of the haemostatic balance is given in Figure I.
Activation of the coagulation cascade is a physiologic response to tissue injury, and
results in the formation of fibrin. Fibrin formation and deposition are a fundamental
process in the repair of injured tissue and protect the body against haemorrhage. Fibrin
is formed from fibrinogen, a reaction catalyzed by thrombin (!Ia) which is generated
after contact activation or tissue activation of a cascade of proteolytic enzymes. The
process of coagulation is controlled by coagulation inhibitors of which antithrombin !II
(AT-lll) and protein Care the most important.
Normal fibrin formation ar.d deposition are temporary reactions, which are curtailed or
reversed when the initiating stimulus is removed, in order to restore normal tissue
structure and function . Fibrin deposition is counterbalanced by the fibrinolytic system
that lyses both fibrin and fibrinogen. Plasmin is the enzyme responsible for lysis of
fibrinogen and fibrin. Plasmin is formed through activation of plasminogen by
activators, such as tissue-type plasminogen activator (t-PA) and urokinase-type
plasminogen activator (u-PA). Fibrinolysis is controlled by inhibitors of the
plasminogen activators (PAl) and inhibitors of plasmin. such as a,-antiplasmin, a,macroglobulin and C1-inactivator.
Disruption of the haemostatic balance may occur when activation of coagulation is
insufficiently counteracted by coagulation inhibitors or when reactive fibrinolysis is
inhibited. Under those circumstances, patients are exposed to the risk of
thromboembolism or multiple organ failure due to widespread intravascular deposition
of fibrin. Consumption coagulopathy. due to increased turnover of plasma coagulation
factors may, after a phase of hypercoagulability. change into hypocoagulopathy and
haemorrhagic diathesis. Consumption coagulopathy may exist alone, it may be
accompanied by slight secondary fibrinolysis or the secondary hyperfibrinolysis may
predominate. The hyperfibrinolysis is a consequence of an excessive production of
9
plasmin with degradation of fibrin, fibrinogen and of coagulation factors and the
appearance of fibrin and fibrinogen degradation products, which in turn inhibit fibrin
polymerization and promote a haemorrhagic diathesis. Finally, lysis of fibrinogen
(primary fibrinolysis of fibrinogenolysis) may occur alone, leading to a haemorrhagic
state through formation of fibrinogen degradation products which have anticoagulant
properties.
Analysis of the haemostatic balance in patients at risk for thromboembolism or bleeding
is essential when treatment is considered. Disturbances of the haemostatic balance will
be reflected by the concentration and composition of the products of the two opposing
processes of coagulation and fibrinolysis - fibrin and fibrinogen derivatives - which can
be used as markers of the status of the haemostatic balance.
With improved understanding of the basic mechanisms of thrombosis and haemostasis,
there has been a rapid development of techniques to assess the haemostatic system and
its associated disorders. One of the newer systems that has recently become available
for the assessment of haemostatic disorders. both thrombotic and haemorrhagic, is the
determination of fibrin degradation products (FbDP) and fibrinogen degradation
products (FgDP) in plasma by enzyme immuno assays (EIA's) based on monoclonal
antibodies.
This thesis comprises studies with monoclonal antibody-based plasma assays for
derivatives of fibrin and fibrinogen in patients with diseases and conditions
characterized by an activated state of coagulation and fibrinolysis.
The aims of the study were:
I. To review the clinical relevance of existing serum-based assays for
fibrin/fibrinogen degradation products in comparison with the newly developed
plasma-based assays for fibrin degradation products (FbDP, D-dimer). for the
total of fibrin and fibrinogen degradation products (TOP) and for fibrinogen
degradation products (FgDP), on the basis of published studies.
10
contact activation
;rrpcekallikcein
XII Xlla\ ----+ activation
---""""1 inhibition
--- -~ destruction
XI a'\ XI AT-Ill
~ IX !Xa '\
) tissue
_ activation VIla
~ X Xa '\
~~~·~ olla (Va II I~
VIII I V ~J~~do~ibno . C protem Ca protetn
Xlllla a2-macr~lobulin C1-LnactLvator
~~ protem s r a,-antiplasmin J
fibrmogen fibrm
Figure L
r---------FgDP FbDP
plasmin -4----- plasminogen 11-- HRG
r t-PA If--- PAl u-P A
i..,.______ kallikrein~ pro-urokinase 1 XI! a ~XII
preka!!ikrein
Schematic presentation of the coagulation cascade and the fibrinolytic
system. AT-III= antithrombin Ill; t-PA =tissue-type plasminogen activator; u-P A=
urokinase-type plasminogen activator; PAl= plasminogen activator inhibitor; HRG =
histidine-rich glycoprotein: FbDP = fibrin degradation products: FgDP = fibrinogen
degradation products.
11
2. To evaluate the assays for D-dimer, FbDP, FgDP and TDP in healthy volunteers
and patients with an activated state of coagulation.
3. To study the haemostatic balance in diseases or conditions associated with an
activated state of coagulation:
a) patients with a suspected or proven thrombotic process.
b) patients with cirrhosis of the liver to see whether haemostatic imbalance
is due to primary fibrinolysis or diffuse intravascular coagulation.
c) patients with septic shock to see whether haemostatic complications are
due to increased activation of coagulation or inhibition of fibrinolysis.
d) patients with an increased tendency to develop thrombotic complications
who are treated with heparin, using measurements of fibrin degradation
products, in order to evaluate effectiveness of treatment.
4. To evaluate the diagnostic value of the D-dimer assay for deep venous
thrombosis using receiver operating characteristic methodology and Baysian
analysis.
12
Chapter 1
MONOCLONAL ANTIBODY-BASED PLASMA ASSAYS FOR FIBRIN(OGEN)
AND DERIVATIVES, AND THEIR CLINICAL RELEVANCE
H. Kroneman12, W. Nieuwenhuizen3, E.A.R. Knot12
Department of Internal Medicine Il1, University Hospital Rotterdam-Dijkzigt, Center
for Clinical Decision Analysis', Erasmus University Rotterdam, Gaubius Institute TN03,
Leiden, The Netherlands
Blood, Coagulation and Fibrinolysis 1991; I :91-111
ABSTRACT
The haemostatic balance can basically be described as the equilibrium between fibrin
formation (coagulation) and fibrin lysis (fibrinolysis). Therefore the status of this
balance may be reflected by the products of these two processes.
Until recently, the tests for assessment of fibrin(ogen) degradation products were
performed in serum since they were based on polyclonal antibodies, which crossreact
with fibrinogen. The use of serum introduces many artefacts and therefore the utility
of these serum tests is limited.
New assays have now become available which can be divided in quantitative enzyme
immunoassays (EIA's) and semiquantitative latex agglutination assays. The new assays
can be carried out in plasma since they use highly specific monoclonal antibodies, the
majority of which do not crossreact with fibrinogen. This makes it possible to avoid the
serum artefacts. Funhermore, these plasma assays can discriminate between degradation
products of fibrin and those of fibrinogen (FbDP and FgDP, respectively).
The possible clinical utility of the new assays is discussed on the basis of literature data
on the following clinical states: deep venous thrombosis (DVT) and pulmonary
embolism, liver disease and liver transplantation, sickle cell disease, renal diseases,
pregnancy and preeclampsia, disseminated intravascular coagulation (DIC), malignancy,
coronary artery disease and thrombolytic therapy. Fibrinolysis appears to be
accompanied by fibrinogenolysis. Detection offibrin(ogen) derivatives may be used to
rule out DVT; to monitor efficacy of anticoagulant treatment for DVT or DIC, and
reflects severity of renal disease but not renal function. High levels of FgDP's were
found during onhotopic liver transplantation and thrombolytic therapy. Fibrin(ogen)
degradation products can not be used to predict reperfusion following thrombolytic
therapy. The fibrinolytic system remained active during normal and complicated
pregnancy and in patients with malignancies.
The new assays provide valuable information on fibrin( ogen)olysis in several diseases.
More information on the haemostatic balance may be obtained by using these new
assays for fibrin(ogen)olysis products in combination with assays for coagulation
products.
14
INTRODUCTION
The haemostatic balance can be described as the equilibrium between the two opposing
processes coagulation and fibrinolysis. Coagulation will lead to fibrin formation, and
the formed fibrin will be lysed by fibrinolysis. It is conceivable that disturbances in the
haemostatic balance will be reflected by the concentrations and composition of the
products of the two processes i.e. fibrin(ogen) derivatives which can be used as
molecular markers of the status of the haemostatic balance.
The measurement of levels of fibrin( ogen) degradation products has been practiced for
at least two decades using a variety of procedures. 1 Several clinical disorders have been
described that are associated with abnormalities in the degradation of fibrin( ogen).
However. since no clinically useful test was available that could measure the
degradation products of fibrinogen specifically, it was not possible to make a distinction
between the degradation products of fibrin and fibrinogen (FbOP's and FgOP's
respectively).
Elevated levels of fibrin(ogen) degradation products (FOP's) have been detected in
serum in the past and were found in venous thromboembolism and diffuse intravascular
coagulation, but also following trauma, surgery. infection. malignancy and sickle cell
anaemia-'·'·"·" It has been postulated that increased FOP levels may be of value in
confirming the diagnosis of diseases characterized by an activated coagulation system
and that normal FOP levels could exclude the diagnosis-'.!' The observation of an initial
rise and subsequent fall of levels of fibrin( ogen) degradation products following
thrombolytic therapy in patients with acute myocardial infarction, suggested that
measurement of fibrin(ogen) degradation products could also be used to monitor the
effect of thrombolytic therapy 15•16
In patient management, a test for the detection of a disease characterized by an
activated coagulation system is clinically useful when it can rule out (suspected)
patients who do not have the disease (specificity= true negative rate= number of
patients without the disease with a normal test divided by the number of people without
the disease) and does conl1rm diagnosis in case the disease is present (sensitivity= true
positive rate= number of patients with the disease with an abnormal test divided by the
15
number of patients with the disease). However, due to a lack of specificity and
accuracy of the serum assays for FDP assessment, the issue of measuring FDP levels
as markers for pathologic conditions has been confounded. Most of these assays could
only be performed in serum samples since they were based on polyclonal antibodies
to fibrin(ogen) and derivatives. As will be pointed out below, the use of serum samples
is unreliable since it introduces many artefacts. Also, polyclonal antibody-based serum
assays could not discriminate between degradation products of fibrinogen and those of
fibrin. Data based on such serum assays should therefore be interpreted with caution.
New techniques have recently become available for the quantitative determination of
fibrin (FbDP) - as well of fibrinogen degradation products (FgDP). These new assays
are performed on plasma samples (thus avoiding the well known serum artefacts) and
are based on monoclonal antibodies. Since a specific test has now become available to
determine degradation products of fibrinogen, these assays permit a separate
quantification of the extent lysis of fibrin (secondary fibrinolysis) and that of fibrinogen
(primary fibrinolysis or fibrinogenolysis).
Application of the new ·assays as well their possibility to discriminate between
fibrinogen-and fibrin degradation products may be of value in the diagnosis of a disease
and the recognition of an altered state of haemostatic function, e.g. a state of strong
fibrinogenolysis that may be associated with an increased risk of bleeding. This may
have implications for the subsequent therapeutical approach in the clinical condition
under study.
The clinical experience with the new assays is steadily growing. and it is becoming
more and more clear that the current diagnostic concepts may have to be adapted, since
they are based on the questionable serum assays. This review article describes the
formation of fibrin(ogen) .md their derivatives. the available plasma assays for their
assessment and evaluates the clinical utility of the new assays in studies described until
now.
16
FORMATION OF FIBRIN(OGEN) DERIVATIVES
a. Fibrin formation
Fibrinogen is a symmetrical glycoprotein with a molecular weight of 340,000 daltons.
It is composed of six polypeptide chains: 2Aa-, 2 BB- and 2 y-chains, which are kept
together by disulphide bonds. In electron microscopic images of fibrinogen a central
nodule and two larger outer nodules are observed. n The central nodule contains the
amino-terminal ends of all six polypeptide chains and is designated as the E-domain.
The more distal nodules are designated as the D-domains and contain the middle and
carboxyl-terminal parts of the Aa-, BB and y-chains. The D-domains are connected to
theE-domain by relatively thin connecting structures. When the coagulation system is
activated, thrombin is formed. Thrombin will conven the soluble fibrinogen to the
insoluble fibrin through a series of intermediate products."' This process is initiated by
the cleavage of the fibrinopeptides A (FpA) from the amino-terminal ends of the Aa
chains (in the centralE-domain). The formed product is fibrin I (or desAA fibrin). The
Aa-chains after release of FpA, are designated as a-chains. The new amino-terminal
ends of the a-chains are able to interact with complementary sites in fibrinogen and in
other fibrin I molecules. At low concentrations fibrin I will be kept in solution by
complexation with fibrinogen. This is known as soluble fibrin. 18~1 At higher
concentrations the fibrin I molecules will polymerise to a macroscopic gel 32, in which
the subunits will be crosslinked by coagulation factor XII! in the presence of Ca2• •
20
Simultaneously, but more slowly, thrombin can also release the fibrinopeptides B (FpB)
from the BB-chains (in the central E-domain), which results in fibrin !I (desAABB
fibrin) formation. Like fibrin I, fibrin II can aggregate to form a gel, in which the
subunits are crosslinked by factor Xlll in the presence of Ca2'.
The individual steps are summarized in Figure 1.
It is conceivable that fibrin formation, which is not counterbalanced by fibrin lysis, will
result in increased blood levels of soluble fibrin. Soluble fibrin could thus be an
indicator of an impending thrombotic event.
17
"l~-------·· '· '·' nm<,_' """""' '"i" "' _____ XI• Y
1, D, Er
XL-FlBR:UI I X-oliqQIIOr&, D-dimer
---~-~------!ibrin(O<jen) olysis
Figure 1. Possible pathways of the formation of fibrin and fibrinogen derivatives.
b. Formation of fibrinogen degradation products (fibrinogenolysis)
Activation of the fibrinolytic system will result in plasmin formation. Plasmin is not a
fibrin-specific protease. however. and it may also degrade fibrinogen.
The carboxyl-terminal ends of the fibrinogen Au-chains (in the distal D-domains) are
highly sensitive to proteolytic activity. In normal healthy individuals, fibrinogen occurs
in three main molecular forms: A high molecular weight form (HMW) in which both
Au chains are intact; a low molecular weight form (LMW) with one slightly shonened
Au-chain and a form with both Au-chains partly proteolysed (LMW').''-35 All three
forms are clottable. The LMW'-form comprises only 4% of all clottable fibrinogen in
a normal individual. The LMW and LMW' forms possibly result from plasmin activity.
It is conceivable that changes in the blood proteolytic activity will first be reflected in
an increase in the LMW' !HMW ratio.
18
When the plasmin digestion of fibrinogen proceeds further, a product is formed, in
which only about 30% of the length of both Au-chains is left. These products still
contain the FpA, and are called fragments X. Subsequently fragment X is cleaved into
one fragment Y and one D-fragment. Fragment Y is composed of the centralE-domain
of fibrinogen (with the FpA) and one D-domain. Eventually, fragment Y will be
cleaved in one E- and one D-fragment. Thus one fibrinogen molecule will eventually
yield one fragment E, and two fragments D, in full agreement with the three nodules
seen in the electron microscopic images.
c. Formation of fibrin degradation products
The degradation by plasmin of the different forms of fibrin mentioned above, proceeds
through intermediate products analogous to those described for fibrinogen. The reason
for this appears to be that plasmin follows the same interdomainal cleavage pattern in
fibrin as in fibrinogen.
Non-crosslinked fibrin I will be cleaved subsequently into fragments X1, Y1, D and E1;
non-crosslinked fibrin ll will yield X11, Y 11, D and E11• The suffixes I and ll denote the
absence of FpA, and of FpA and FpB. respectively.
The crosslinked fibrins I and II consist of very long polymers of fibrin I and II. in
which the subunits are crosslinked by isopeptide bonds. The subunits in these polymers
will be proteolysed by plasmin in a random fashion. This will result in smaller (soluble)
fragments of the original polymers. with a range of molecular weights?'·" These are
collectively designated as X-oligomers."-'1.36 Upon prolonged digestion by plasmin
these X-oligomers will be convened to fragments D-dimer i.e. two covalently bound
D-domains 20'36 and fragments E.
d. Molecular features of different fibrin(ogen) derivatives
On the basis of the brief description of the various fibrin(ogen) derivatives in the
preceding paragraphs. several molecular features can be recognized, which discriminate
one (group of) fibrin(ogen) derivative(s) from another. These are summarized in Table
L
19
Table 1 Molecular characteristics of fibrinogen and some fibrin(ogen) derivatives.
fibrin(ogen) derivative
HMW-Iibrinogen
soluble fibrin/non-crosslinked
crosslinked fibrin
molecular characteristics
carboxyl-terminals Acx-chains intact;
contains FpA; no crosslinks
contains no FpA or crosslinks; new
amino-terminus in a-chains
contains no FpA; contains cross-
links; new ex-chains amino-terminus
fibrinogen degradation products (FgOP) contains FpA; altered conformation
as compared with fibrinogen (neo-
non-crosslinked fibrin degradation
products (FbDP)
crosslinked fibrin degradation
products (XL-FbDP)
epitopes)
contain no FpA; altered conformation
as compared with fibrin (nee-
epitopes)
contain no FpA; altered conformation
as compared with fibrin ( neo
epitopes); contain crosslinks.
Table I Molecular characrerisctics of fibrinogen and some fibrin(ogen) derivatives
As will be shown below n~w monoclonal antibody-based assays are mostly based on
the specific recognition of these molecular features.
20
ASSAYS FOR FIBRINOGEN AND ITS DERIVATIVES
Serum samples should not be used
Even in case of a disease, the blood concentrations of fibrin(ogen) derivatives will be
low, as compared with the fibrinogen concentration i.e. in the microgram/ml range. For
that reason sensitive immunological methods such as enzyme immunoassays (EIA) or
latex agglutination assays are required. Until relatively recently, only serum could be
used as a sample since the available polyclonal antibodies crossreact virtually
completely with fibrinogen. The latter had to be removed e.g. by serum preparation.
Well-known serum assays are the Thrombo-Wellco assay and the tanned red-cell
hemagglutination inhibition assay.37 Serum, however, is a notorious source of anefactual
results:
*
*
*
•
Incomplete clotting of the cross-reacting fibrinogen may occur in cases of
dysfibrinogenemia: when a patient has been exposed to heparin 38 or when
anticoagulant fibrinogen degradation products are present.39-41 In polyclonal
antibody-based serum assays this will inevitably lead to too high or false
positive results. Those can also result from partial lysis of the clot during serum
preparation. not only in hyperfibrinolytic patients. but even in normal
individuals. 19
Some degradation products will coagulate ' 0•42 or become adsorbed to the
clot."·''·" For that reason they will not be recovered in the serum and false
negative or too low results may be found.
During serum preparation fibrinogen degradation products will loose their FpA.
As a result they can no longer be discriminated from non-crosslinked fibrin
degradation products, and primary fibrinogenolysis can not be detected.
Serum can obviously not be used for the assessment of the products of ongoing
coagulation in a patient.
These problems inherent in the use of serum could be avoided by the use of plasma.
Only relatively recently. with the advent of the monoclonal antibody technology, it has
2l
become possible to develop assays which can be performed with plasma samples and
thus avoid the serum artefacts.
The new assavs
In the following paragraph a brief outline will be given of the currently available assays
based on monoclonal antibodies, which can be performed with plasma samples. The
fundamental change as compared with the old technology is that monoclonal antibodies
are used. which do not react with fibrinogen (except in the assay for intact fibrinogen,
which will be described). This allows the assessment of coagulation and
fibrin(ogen)olysis products in the same individual sample.
Available monoclonal antibody based plasma assays
Enzyme immunoassays (E!A)
All the available enzyme immunoassays are of the sandwich-type. They are based upon
a specific monoclonal antibody absorbed to the wells of a microtitration plate. This is
the solid phase. or capture antibody. Usually the capture antibody is directed against
one of the molecular characteristics of a fibrin(ogen) derivative, summarized in Table
1. In the second step, the captured fibrin(ogen) derivative is tagged with an enzyme
labeled panspecific monoclonal antibody (or polyclonal antiserum), or with another
enzyme-labeled monoclonal antibody with a defined specificity. The enzyme is usually
horseradish peroxidase (HRP).
The most relevant EIA • s described today are summarized in Table 2. As can be seen
plasma assays are now available (or will be available soon) for intact fibrinogen;
soluble fibrin; the total of degradation products derived from fibrinogen and fibrin; for
22
Tabla 2 Currendy described EtA's for fibrinogen and its derivates
specificity of
EIA specific for
Intact fibrinogen
soluble fibrin
fibrin.Q.9£!.! degradation products (FgOP)
X--¢ligomors
0-.dimer
0--dimer
fibrin degradation products (FbDP)
total of degradation products (TOP) of fibrin (FbDP) plus those of fibrinogen (FgOP)
Capture antibody
Intact carboxyHermlnus of Auchains (GS)
now amino-terminus of fibrin achain (anti-lbn-17)
altered conformation In fibrin{ogon) degradation products (FOP-14)
X--¢/igomers (NiBn 52)
crosslinks In 0-dimer {DD3B6)
mainly for 0-dimor
altered conformation in fibrin(ogon) degradation products (FDP-14)
altered conformation in fibrin{ogon) degradation products (FOP-14)
•.u.l from Organon Teknika, Turnhout, Belgium
tagging antibody
amino-terminus of Aa-cha!n, Incl. FpA (VIS)
non-specllic (pofyc!onal)
amino-terminus of Aa-chain (incl. FpA) (Y18)
X--¢1igomers (NiBn 178)
panspecific anti-0 monoclonal antibody
panspecilic antibody
monoclonal antibody against D of non-crosslinked fibrin and 0-dimer (OD-13)
mixture of two monoctonals (Y18 + DD-13) {see above)
l Maboo, Brisbano, Australia, and America! Diagoostica, Greenwich, USA. Diagnoslica Stage, Asni9res, France: Boehringer Mannhelm, Mannheim, FRG
remarks
Commercial!y available soon\ Duration of assay onfy 45 minutes
Commercially availablo2• Duration
c>l assay only 45 minutes
Commercial!y available,.
Commercially available~.
Commercially available1• Duration
of assay only 45 minutes
Commercially available'. Duration of assay only 45 minutes
referoooos ----21,22
23,24,25
26,27
10,11 ~
28,29
30,27
19
early fibrin degradation products (X-oligomers); for late fibrin degradation products,
and for fibrinogen degradation products. With this panel of E!A' s it is possible to draw
conclusions about the status of the haemostatic balance in much detail.45
Agglutination assays
Apart from the quantitative EIA's described above, also semiquantitative latex
agglutination assays are available for use in plasma. These are:
* An assay for D-dimer, based on DD3B6 28.29 monoclonal antibody (see table)
and available form MabCo, Brisbane, Australia; from American Diagnostica,
Greenwich, USA and from Ortho Diagnostics. Beerse, Belgium.
*
*
A D-dimer test from Diagnostica Stago. Asnieres, France; from Boehringer
Mannheim. Mannheim, FRG and from Organon Teknika, Turnhout, Belgium.
A D-dimer test reported by Soria 46• which is not commercially available.
CLINICAL STUDIES WITH THE NEW ASSAYS
The clinical experience of the new assays is rapidly growing and degradation products
of fibrin and fibrinogen were measured in plasma samples of patients with several
different diseases. Theoretically, clinical disorders associated with elevated fibrin(ogen)
degradation products. as assessed by the new assays. can be divided in three different
categories:
A) diseases in which the process of primary fibrinolysis (fibrinogenolysis) dominates,
resulting in elevated levels of FgDP's;
B) disorders with a pronounced increase of FbDP levels. indicative for the
occurrence of secondary fibrinolysis: and
C) clinical states characterized by an enhancement of both fibrinogenolysis and
fibrinolysis.
Most diseases with increas~ levels offibrin(ogen) degradation products can be placed
in this last category and we prefer to discuss the different diseases separately.
24
Deep venous thrombosis and pulmonary embolism
The diagnosis of deep venous thrombosis (DVT) based on clinical symptoms only, is
missed in 50% of the cases 47·" Therefore it should be confirmed by objective methods
and of these, ascending venography is considered as a golden standard. There is a need
for non-invasive tests like impedance plethysmography, echography and also laboratory
tests. A simple screening test that selects patients for further diagnostic procedures can
be clinically useful. Efficacy of heparin treatment for DVT is often evaluated by
repeated venography. Since this procedure is time consuming, expensive and not
without risks, a method is needed that could replace this invasive technique. Using the
new assays, the levels of fibrin(ogen) degradations products levels were measured in
plasma to determine their role in both diagnosis and evaluation of treaunent for DVT.
Using a semiquantitative latex assay for detection of plasma D-dimer it was possible
to make a clear distinction between normal subjects and hospitalized patients with DVT
confirmed by venography." In that study, all patients with confirmed DVT had plasma
D-dimer levels above the r.oinimal detectable concentration of 250 ng/ml. None of the
healthy controls reached these levels. However, another study showed that plasma D
dimer levels of hospitalized patients with DVT (the authors do not mention how the
diagnosis was confirmed) showed some overlap (<5%) with those of healthy
volunteers. 50 In both studies. the correlation between plasma D-dimer levels assayed by
the semiquantitative latex assay and the quantitative EIA was highly significant
(r=0.83.P<0.001 ' 9; r=0.82. P<O.OOOl 50
) The plasma D-dimer latex assay was also
applied to a population consisting of hospitalized-and outpatients suspected for DVT.51
In this study the sensitivity and specificity of the latex assay (cut-off value for test
positivity: 200 ng/ml) for DVT were 73% and 69% respectively. The test results were
not separately determined in the in- and outpatient groups. The diagnosis of DVT was
made by ascending venography. To be useful as a screening test for DVT, the latex D
dimer assay must allow to discriminate in the group of patients clinically suspected for
DVT those who have DVT from those who don't. This can be determined by
calculating the predictive value of a positive or negative test result, meaning the
probability of having or not having the disease.
25
In that study 51 with a prevalence of DVT of 72%, the positive and negative predictive
value were 63% and 78% respectively. In another study two different D-dimer latex
assays were used in hospitalized -and outpatients suspected for DVT 52 The prevalence
of DVT, as assessed by venography was 65%. One of the assays using a cut-off value
of 500 ng/ml had a sensitivity of 76%, a specificity of 87% and the positive and
negative predictive values were 80% and 85% respectively. The other latex assay (cut
off value 200 ng/ml) had a sensitivity, specificity and a positive and negative predictive
values of 60%, 97%, 92% and 80% respectively. The test results were calculated for
the whole population and not for the subgroups of hospitalized patients and outpatients
separately. Similar observations were reported by others."
The above studies mentioned above clearly demonstrate that D-dimer levels, as
determined by the latex agglutination assay, cannot be used to exclude the diagnosis
DVT. Using this assay as screening test for DVT the diagnosis will be missed in 15%-
22% of the patients 51 .s2 which is an unacceptably high percentage. Based on these
grounds the conclusion is justified that the D-dimer latex assay is unsuitable as a
screening test for DVT.
Using a D-dimer EIA, it was shown the level of plasma D-dimer was elevated in
hospitalized patients with DVT, confirmed by venography."·" No overlap was
demonstrated with the D-dimer levels in normal healthy controls (cut-offvalue for test
positivity:200 ng/ml).
In a population consisted of hospitalized patients suspected for DVT, the D-dimer EIA
had a sensitivity and specificity for DVT of I 00% and 23% respectively, as assessed
by venography." The used cut-off value for test positivity was 250 ng/ml. The
prevalence for DVT was 52% and the negative predictive value was !00%. Similar
observations were made in other studies using the D-dimer EIA 's.51 .52 The studies
mentioned above show that the diagnosis of DVT can be ruled out in hospitalized
patients with clinically suspected DVT in case of normal D-dimer levels, as determined
by the quantitative D-dimer EIA. As a consequence ascending venography could have
been avoided in these patients.
The ElA for plasma D-dimer was also applied to an outpatient population with
26
clinically suspected DVT.5" The sensitivity and specificity of the D-dimer ElA (cut-off
value:200 ng/ml) for DVT were 100% and 61% respectively. as evidenced by
venography. In that population with a prevalence for DVT of 35%. the predictive value
of a positive test was 61% and the negative predictive value was 100%.56 Similar
results were found in other studies.51·52 One of these studies showed that the negative
predictive value declined from 100% to 93% when the cut-off value for test positivity
was raised from 200 ng/ml to 500 ng/ml.52 Another study reported a negative predictive
value of 98% using a cut-off value of 200 ng/ml.53 These studies. which were
performed in outpatients. confirm the observations obtained on hospitalized patients.
that normal D-dimer levels. determined by EIA. can be used to exclude the diagnosis
in patients with suspected DVT. Based on these results it was suggested that the D
dimer EIA could be used as a screening test for DVT and that ascending venography
can be avoided in patients with D-dimer levels below the cut-off point for test
positivity.53·54 In clinical practice this would be a great advantage. However, the
predictive value of a test can dramatically be affected by the prevalence of the disease
in the studied population i.e. the positive predictive value decreases with the prevalence.
but the negative predictive value may rise simultaneously. It is not advisable to apply
the predictive value of a test obtained in one population to another without knowing the
prevalences in these populations. The utility of the EIA for D-dimer as a screening test
for DVT in a certain population can be determined by studying the costs-effectiveness
in relation to the prevalence of DVT in that population.
With a specific EIA for fibrin degradation products (FbDP"s). increased plasma levels
of FbDP"s were demonstrated in patients with DVT relative to FbDP levels in plasma
of healthy controls.30~'·" Patients with DVT had also increased levels of fibrinogen
degradation products (FgDP). although the latter were less elevated than the FbDP"s.57
This suggests that secondary fibrinolysis was accompanied by fibrinogenolysis.
Applied to outpatients. clinically suspected for DVT. the EIA for FbDP (cur-off value
270 ng/ml) had a sensitivity of 92% and a specificity of 20% for the diagnosis of DVT.
as determined by impedance plethysmography (!PG).57 A second lPG examination was
performed after one week when the first lPG examination was negative. In the same
27
study, the sensitivities of the E!A for FgDP (cut-off value: 250 ng/ml) and the EIA for
TDP (cut-off value: 590 ng/ml) were 90% and 95% respectively. The specificities were
25% and 16% respectively. In this population with a prevalence of DVT of 25%, the
negative predictive values of the ElA's for TDP, FbDP and FgDP were 90%, 87% and
88% respectively. The test characteristics of the EIA's were better, if compared to the
results of the first lPG examination only, e.g. the sensitivity and the negative predictive
value of the E!A for TDP were 98% and 97% respectively. The utility of these EIA's
as screening test for DVT can be determined in the same way as described above.
Until now, no studies have. been described that investigated the test characteristics of
the EIA's for TDP. FbDP and FgDP in hospitalized patients suspected for DVT.
All patients with pulmonary embolism (PE), confirmed by lung perfusion scanning had
increased plasma levels of D-dimer when compared to the levels of D-dimer in normal
controls, as detected by ElA55 This finding was confirmed in another study where all
patients with PE, as assessed by pulmonary angiography, had elevated levels of plasma
D-dimer (above the normal value in healthy volunteers of 130 ± 65 ng/ml) in contrast
to the normal level of D-dimer that was observed in one patient in whom PE could be
excluded.60 The same results were found using an EIA, specific for X-oligomers (cross
linked fibrin degradation products with a high molecular weight). 61
With regard to daily clinical practice it is important to know whether the level of
plasma D-dimer can predict the presence or absence of PE in a population of patients
suspected for PE. By examination of patients suspected for PE using an EIA for D
dimer, it was demonstrated that a high suspicion of PE on the basis of a lung scan
corresponded to D-dimer levels above 500 ng/ml in all patients, but in case the scan
was inconclusive, a D-dimer value below 500 ng/ml could not exclude PE.62
Plasma of patients suspecred for PE was also assayed with an E!A that recognizes
cross-linked D-dimer fragments and related high molecular weight fibrin derivatives
containing D-dimer (XDP's).63 Using a cut-off value for test positivity of 145 ng/ml,
the sensitivity and specificity of that EIA for the diagnosis of PE were 89% and 44%
respectively, as evidenced by lung perfusion scanning. In that population with a DVT
prevalence of 38%, the predictive value of a negative test was 92%. This study
28
demonstrates that the diagnosis PEcan not be excluded in patients clinically suspected
for PE with normal levels of XDP, as assessed by this XDP EIA. Since the diagnosis
of this potentially life treathening disease will be underestimated in 8% of the cases 63
it can be concluded that the EIA for XDP assessment alone, can not be used as a
screening test for PE in clinically suspected patients.
Similar observations were found with the semiquantitative latex agglutination assays
for D-dimer. An increased plasma D-dimer level was demonstrated in 98% of patients
with PE confirmed by lungscan but in none of the healthy controls (cut-off level for
test positivity: 250 ng/ml).'9 The observed overlap between D-dimer levels in patients
with PE and normal controls was confirmed by others using a cut-off level for test
positivity of320 ng/ml. 50 Another study showed that 54% of the patients with clinically
suspected PE. who had a high probability on lungscan, had a negative latex
agglutination assay (cut-off value: 500 ng/ml).64 This supports the finding obtained by
application of the D-dimer EIA "s that it not possible to exclude the diagnosis PE in
patients with clinically suspected PE on the basis of a normal D-dimer level alone. In
these patients. further diagnostic procedures are indicated.
The possibility to use plasma levels of FbDP or D-dimer, as markers of secondary
fibrinolysis to monitor thrombolytic and heparin therapy in patients with venous
thromboembolism has been investigated using EIA 's.58•59
•65 It was demonstrated that a
change in plasma FbDP level was reflected in a change of thrombus size during heparin
treatment. This was assessed by repeated venography based on Marder scores. A fall
in FbDP level was accompanied by thrombus regression. but when the FbDP level
remained constant during treatment. no reduction of thrombus size could be observed. 58
In that study a good correlation (r=0.79) was found between high initial plasma FbDP's
and thrombus size. These results suggest that FbDP detection by EIA is useful in the
monitoring of heparin treatment. More data are needed to confirm these findings.
Thrombolytic therapy with streptokinase (SK) or urokinase (UK) was followed by an
increase of FbDP's during the first 12 hours of treatment.59•65 Following SK or UK
therapy it was demonstrared that low initial plasma FbDP level predicted a poor
decrease in thrombus size, but the changes ofFbDP levels were not correlated (r=0.21)
29
to the changes in thrombus size. as assessed by repeated venography." This was
confirmed by others65 However, an increase of FbDP levels is no evidence for lysis of
the thrombus, since elevated levels of FbDP were also found in normal healthy
volunteers following t-PA infusion.66
Liver disease and liver transplantation
Patients with chronic liver disease often develop localized bleeding episodes (ruptured
esophageal varix, haemorrhagic gastritis 67) that tend to massive haemorrhage and are
an important cause of death in these patients."·" The disturbance of blood coagulation
is complex in patients with liver cirrhosis. It has been hypothesized that liver cirrhosis
is frequently complicated by disseminated intravascular coagulation (DIC) and that
consumption coagulopathy due to DIC plays a role in the pathogenesis of haemorrhage
in patients with liver cirrhusis.70.75 However, this view is not accepted by all.n.76
·77
A number of studies reponed that fibrinolysis is enhanced in patients with liver
disease-''·" This was evidenced by the observation of increased plasma t-PA antigen
levels presumably due to ' diminished hepatic clearance of t-PA 81•84 and decreased
levels of a,-antiplasmin probably due to a decrease in synthesis capacity of the affected
liver.79•80
The new plasma assays have been applied to patients with liver cirrhosis and patients
undergoing liver transplantation. Using an ElA for intact fibrinogen (HMW + LMW)
in patients with liver cirrhosis, classified according to the Child-Pugh classification, it
was shown that the level of intact fibrinogen decreased with increasing severity of the
disease, and was lowest in Child C."' The FgDP levels. as detected by EIA, were only
slightly elevated compared to the level in normal controls and did not correlate to the
severity of the disease.
This either suggests that extensive fibrinogenolysis is not a common feature in patients
with liver cirrhosis and the decreased levels of intact fibrinogen can be explained by
a lowered synthetic capacity of the diseased liver or that fibrinogen beyond the stage
of LMW' formation does not occur (see above). Further it was shown in that study that
both the FbDP and TDP levels were elevated and showed a positive correlation with
30
the severity of the disease. This indicates that the process of secondary fibrinolysis is
more active in patients with an advanced stage of liver cirrhosis.
The observation of increased FbDP levels, as detected by EIA, in patients with liver
cirrhosis was confirmed in another study.86 These findings may be indicative of the
presence of a hypercoagulable state, although a failure of the clearance of fibrin
degradation products by the damaged liver can not be excluded. Further studies are
needed to answer this clearance question and to determine whether increased FbDP
levels are correlated with organ failure due to fibrin deposition.
Using a semiquantitative D-dimer latex assay in patients with liver cirrhosis classified
according to Child and Pugh it was shown that the level of D-dimer was elevated (p
< 0.001) in patients with Child C classification as compared to the D-dimer levels in
patients with Child A and B.87 In that study. no significant difference was observed
between the D-dimer level in patients with Child A and B classification. Elevated
plasma levels of D-dimer, detected by latex assay, were also demonstrated in patients
with liver disease (the authors did not specify the disease) and increased when DIC was
superimposed.88 These latter studies, using semiquantitative assays. suppon the previous
conclusion based on EIA's that secondary fibrinolysis is more pronounced in patients
with advanced liver cirrhosis.
Orthotopic liver transplantation (OLT) is accompanied by serious bleeding
complications 89•90 presumably caused by a hyperfibrinolytic state 91
•92 or secondary to
diffuse intravascular coagulation (DIC).93'95 Measurement of plasma levels of
fibrin(ogen) degradation products by EIA's, showed that the levels of FbDP's and
FgDP's increased significantly relative to pre-operative values, at the end of the
anhepatic phase during OLT. The rise of FgDP levels was, however, less than the rise
of FbDP levels.96 In that study. FgDP levels reached a maximum shortly after
recirculation of the gf'<~ft and just before the FbDP peale Similar findings were reponed
by others86•97 Using a D-dimer EIA. it was demonstrated that plasma D-dimer level also
10se in the anhepatic phase and following hepatic revascularization during OLT.9'-100
In conclusion, the rise in FgDP level clearly demonstrates the occurrence of
fibrinogenolysis during Ol.T, which may considerably contribute to the disturbance of
31
the haemostatic function. It can be concluded that OL T is accompanied by both
hyperfibrinolysis and an activated state of coagulation. The observed peaks of the FbDP
and FgDP levels in the early post anhepatic phase correspond to the clinical picture of
increased bleeding tendency during that period. Whether the increase in the levels of
fibrin( ogen) degradation products are indicative for the intra-operative blood loss and
can be used to identify patients with active fibrinolysis with an increased risk of life
treathening haemorrhage. remains to be determined.
During auxiliary partial liver transplantation (APLT) ElA • s were also used to measure
fibrin(ogen) degradation products. It was shown that FgDP, as detected by an E!A, was
elevated during the operation (maximum level: 1.0 ug/ml) but was no longer detectable
during the first day after the operation, whereas TDP and FbDP showed no abrupt
changes during the operation. FbDP, however, remained detectable until 25 days after
transplantation. 101 These data are based on measurements in one patient and more
studies are needed to confirm these findings.
Sickle cell disease
The primary biochemical lesion in sickle cell disease is an alteration of the structure
in the B chain of the haemoglobin resulting in the so called haemoglobin S. Red cells
containing haemoglobin S at a high concentration become eventually irreversibly
sickled. and this has two main effects. Firstly it shortens the survival of the erythrocytes
and secondly the viscosity of the blood increases because of the formation of
aggregates of the sickled cells which may result in stasis and hence occlusion of the
microvasculature. which is experienced as painful by many patients.
It is not clear whether activation of the coagulation system contributes to the vasa
occlusive crisis in patients with sickle cell disease. Several observations suggest that
enhancement of coagulatic'n with formation of fibrin may play a role.'·102 Fibrin has
been detected in vessels ocduded by sickle cells. 103 During steady state, the fibrin( ogen)
fragment E level was elevated' Fibrinogen survival is shorter than normal in the steady
state of the disease, and is further shortened during painful crisis. 1"' Many coagulation
laboratory tests show abnormalities during painful crisis 105•106 but no test is available
32
that clearly detects the occurrence of vase-occlusive crisis. Fibrin degradation products
were measured in plasma of patients with sickle cell disease during steady state and
painful crisis as markers of fibrinolytic activity.
Using a semiquantitative D-dimer latex assay with a minimum detectable concentration
I ug/ml. it was demonstrated, that 10% of the patients with sickle cell disease who
were asymptomatic, had elevated D-dimer levels as compared to normal healthy
controls.107 The same study also showed that 97% of the hospitalized symptomatic
patients during painful crisis had elevated D-dimer levels, but D-dimer was not
detectable in plasma samples of hospitalized patients that had become asymptomatic
following a painful crisis. With respect to the clinical status of the patients (assessed
by history and physical examination) a significant correlation (p < 0.001) between
disease activity and the presence of D-dimer plasma level was noticed. D-dimer was
detectable in only 10% of the steady state (asymptomatic) patients, indicating that the
fibrinolytic activity is more pronounced during painful crisis.
Using a D-dimer E!A, the observation of elevated D-dimer levels during painful crisis
was confirmed in another study .108 1n that study, significant differences (p<O.OOI) in D
Dimer levels were noticed between healthy controls. asymptomatic patients with sickle
cell disease during steady state and symptomatic patients during painful crisis. In 93%
of the cases, D-dimer plasma levels in the asymptomatic patients during steady state
were above the upper limit of the level of healthy individuals (129 ng!ml). This is
higher than the 10% in the previously mentioned study 107 and can be explained by the
fact that the ElA is more sensitive for the detection of low grade activation of
coagulation and reactive fibrinolysis than the latex agglutination assay.
Plasma of patients with sickle cell disease has not yet been assayed by EIA's for TOP,
FbDP and FgDP.
Based on the aforementioned data, it can be concluded that application of the new
assays has demonstrated that fibrin formation and subsequent lysis is not an exclusive
feature of the vase-occlusive crisis, but is also present during steady state. Whether
measurements of plasma levels of fibrin( ogen) degradation products are useful to detect
the occurrence of vase-occlusive crises or other complications in patients with sickle
33
cell disease may be assessed by studying the predictive value of the assays in the
clinically relevant population.
Pregnancy and Cpre)eclampsia
During preeclampsia, a decrease in platelet count, anti-thrombin III (AT-lli),
plasminogen and <X;,-antiplasmin is observed-"'·''' However, none of these laboratory
data reflects coagulation abnormalities accompanying preeclampsia satisfactorily_uz.uJ
Therefore these tests are not useful to monitor the occurrence of preeclampsia in high
risk groups. It has been suggested that measurements of levels of fibrin( ogen)
degradation products may be of value to diagnose (pre)eclampsia since it was observed
that toxemia of pregnancy may be developing when these levels increase.11'-
116
However. the assays were performed in serum.
Plasma of women with normal, uncomplicated pregnancy and of women with
pregnancy complicated by preeclampsia was assayed for D-dimer and fibrin( ogen)
degradation products, using semiquantitative latex agglutination tests. 117.119 Plasma
samples of women with normal pregnancy were negative for fibrin( ogen) degradation
products in 98%-100% of the cases '"·"'and only 5% of the women showed detectable
levels of plasma D-dimer (cut-off value: 200 ng/ml)."' In that study the increase of D
dimer showed a positive correlation (r=0.64) with the fibrin(ogen) degradation products
levels.
During the preeclamptic state. elevated plasma levels of fibrin(ogen) degradation
products have been observed.'"·'" In 38,7% of the preeclamptic women elevated
Plasma D-dimer levels were observed. in contrast to the healthy pregnant women in
whom no elevations were observed (detection limit 2 J.Lg/ml).117 The D-dimer positive
women had a higher blood pressure (systolic and diastolic), more proteinuria and
abnormal liver function rests, a higher serum creatin and blood urea nitrogen than
preeclamptic women with normal D-dimer levels. In addition the D-dimer positive
women showed a higher risk for prematurity and low binh weight. It was concluded
that the semiquantitative latex agglutination assay for D-dimer may be a useful test to
screen preeclamptic women in order to define a subgroup with a high risk for
34
coagulation abnormalities.
Also quantitative E!A's were used to study plasma levels of fibrin(ogen) degradation
products in women with normal and preeclamptic pregnancy_61.1I9-122 In women with
normal pregnancy the plasma levels ofTDP, FbDP, FgDP and D-Dimer were elevated
as compared with a control group of healthy non-pregnant women, and especially
increased during the last nimester of pregnancy.119-123
This is in contrast with the findings obtained with the Thrombo-Wellco serum test that
detects FOP. At no time during normal pregnancy significant increases of the FDP
levels could be demonstrated. 120 This may be explained by the fact that the serum assay
is less sensitive than the EIA, or be an artefact due to serum preparation (see above).
Using an EIA for detection of cross-linked fibrin degradation products (X-oligomers),
it was found that all women with clinically diagnosed preeclampsia had elevated plasma
levels of X-oligomers, and showed no overlap with those of women with uncomplicated
pregnancy. 61
Despite a number of studies that reports a depressed fibrinolytic activity during
pregnancy 124-126
, the studies mentioned above using EIA's, demonstrate an increase of
plasma levels of fibrin degradation products during normal and complicated pregnancy.
This suggests that the fibrinolytic activity is not necessarily impaired during pregnancy.
Detection of plasma fibrin(ogen) degradation products by EIA's are a useful aid in the
early diagnosis of preeclampsia and may help to formulate therapeutic action to prevent
a full blown expression of this condition.
Renal diseases
The role of hemostatic disorders in the pathogenesis of renal vascular and glomerular
diseases is not fully understood.
Glomerular fibrin deposition is observed in many cases of glomerulonephritis and it is
assumed to play a role in glomerular degeneration. 127•128 The presence of fibrin and
fibrinogen (degradation products), was demonstrated immunochemically within the
mesangium, in vascular cells and along the basement membrane in acute homograft
rejection, postpartum renal failure and rapidly progressive glomerular nephritis.129-131
35
It has been shown that FOP, as detected by serum assays, are elevated in several renal
diseases. 132
Since extra-renal sites of fibrin(ogen)olysis may also be responsible for FOP elevation,
it was suggested and confirmed that FDP levels measured in the urine would reflect
disease activity better than FOP levels detected in serum.133-136
An EIA was used to detect urinary fibrin( ogen) degradation products of patients with
various types of primary and secondary chronic nephrititis.137 In that study, urinary
Fb(g)DP levels were significantly higher (p<0.05) in patients with severe types of
proliferative, membranous glomerulonephritis and in the active phase of SLE as
compared to the urinary levels of normal volunteers. No correlation (the correlation
coefficient was not given) was found between the levels of urinary Fb(g)DP and that
of proteinuria.
Urine of patients with glomerulonephritis characterized by active cellular crescents was
examined using an EIA or a radioimmunoassay (RIA) specific for cross-linked fibrin
degradation products (XLFDP's).'"-' 39 The urinary level of XLFOPs correlated weakly
(r=0.52, p<O.OI) with the percentage of glomeruli containing cellular crescents. No
significant correlation (r=0.6i,p=ns) was found between the levels of urinary XLFOP's
and proteinuria.'" This may suggest lysis of fibrin(oid) material within the crescents
or Bowman's space that was demonstrated by percutaneous renal biopsy in a majority
of these patients. 138
In patients with focal segmental sclerosis (FSS) or membranous nephropathy (MN), the
urinary levels of XLFDPs correlated (r=0.73 and 0.70 for FSS and MN respectively)
with the level of proteinuria. 139 The ratio urinary XLFDP/proteinuria was significantly
higher (p < 0.05) in patients with glomerulonephritis (GN) with crescents than in
patients with FFS and MN. The ratio urinary XLFDP/proteinuria was used to exclude
that portion of urinary XLFDP that was filtered from the blood. but whether the
detected high ratio in patients with crescents GN reflects intraglomerular fibrinolysis
remains to be determined.
In children with several kinds of proliferative glomerulonephritis, the demonstrated
increased urinary levels of XLFDPs and D-dimer, as assessed by EIA's, showed no
36
correlation with the histologically detected mesangial proliferation or with the
selectivity of proteinuria. " 0 This may suggest that lysis of intraglomerular fibrin is not
the only source of urinary cross-linked degradation products.
Recently it has been shown that D-dimer levels in plasma, as detected by EIA, were
significantly elevated in non-diabetic chronic renal failure (p < 0.05), diabetic
nephropathy (p < 0.01) and acute renal failure (p < 0.01) when compared to the D
dimer plasma levels of healthy controls(< 220 ng!m1)."1 In that study, no correlation
(r = -0.14) was found between plasma D-dimer and creatinin clearance. This suggests
that detection of D-dimer in plasma by EIA can be used a marker of fibrin breakdown
in several renal diseases and reflects disease activity. More studies are needed to
confirm these findings. We are not aware of studies that used semiquantitative D-dimer
assays or EIA 's for TDP. FbDP or FgDP assessment in renal disease.
It can be concluded that EIA 's specific for the detection of urinary cross-linked fibrin
degradation products can be used to assess activity of several renal diseases. The EIA 's
are not suitable to assess renal function itself. since none of these assays showed
significant correlations between fibrin degradation products and creatinin clearance. The
origin of the observed increased levels of urinary cross-linked fibrin degradation
products in patients with several renal diseases is not fully clear: it may result from
filtration from the blood (because of an increased permeability of the damaged
basement membrane). or from lysis of fibrin in the lower urinary tract or the kidney.
In the future, the new assays performed in plasma may be applied to patients with
homograft transplants, since in the past it has been shown that a rise in serum FDP is
associated with impending rejection of the graft.142
Disseminated intravascular coagulation ( D!C)
DIC may lead to a dramatic and generalized haemostatic failure with a wide spectrum
of clinical symptoms and is often associated with a variety of other pathological
conditions. 12•143
-148 In case of a dominant intravascular clotting process and minimal
secondary fibrin(ogeno)lysis, DIC may present primarily as diffuse thrombosis. When
the process of secondary fibrinolysis dominates, the clinical manifestation shifts towards
37
haemorrhage. Patients suspected of having D!C often demonstrate both clinical
conditions simultaneously, and a laboratory test capable of discriminating between these
two situations is needed since both clinical conditions require a different therapeutic
approach.
However, in patients with a clinical condition indicative of D!C, the currently used
haemostasis parameters like prothrombine time, thrombin time, activated partial
thromboplastin time, factor V, fibrinogen, platelet count and antithrombin Ill become
difficult to interpret and none of these tests is specific for the diagnosis or the
monitoring of therapy in case of D!C.149.152
Plasma of patients suspected of having DIC was examined by using latex assay's for
Fibrin(ogen) degradation products.49.so.s7.147.153-1SS
It was shown that all patients with DIC had elevated levels of plasma D-dimer as
compared to D-dimer levels in normal healthy controls." The minimal concentration
detectable of the used latex assay was 200 ng!ml. In that study, the diagnosis of DIC
was based on a prolonged thrombin time. hypofibrinogenaemia, thrombocytopenia and
elevated serum FDP's (>8 )J.g/ml) assessed by the Thrombo-Wellcotest in combination
with clinical conditions known to be associated with DIC. This finding was confirmed
in another study using a latex assay with a minimal detectable concentration of 250
ng!ml."
The latex assay was also vsed to measure plasma D-dimer levels in patients clinically
suspected of having DIC. 15' In that study. the presence of fragment D-dimer, as
demonstrated by immunoblotting. was used as a criterion for the diagnosis of DIC and
all patients who had fragment D-dimer present showed clinical conditions consistent
with the diagnosis of DIC. The sensitivity and specificity of the D-dimer latex assay
for the diagnosis of DIC. as determined by immunoblotting, was 85% and 97%
respectively. The prevalence was 51% and the positive and negative predictive value
was 97% and 86% respectively. Although varying of the cut-off point may change the
test results, the aforementioned study demonstrates, that the negative D-dimer latex
assay alone can not be used to exclude the diagnosis of DIC in patients suspected of
having DIC. More studies ·tre needed to confirm the test characteristics of the D-dimer
38
latex assay obtained in this study. D-dimer measurements by latex assay may be
valuable in addition 10 the currently used laboratory tests in patients suspected of
having DIC.
Plasma of patients with DIC was also assayed by quantitative EIA's to demonstrate
elevated levels of fibrin(ogen) degradations products.26·30.so.ss.Gt,t22.IS6-tGo using an EIA
specific for cross-linked fi!:Jrin degradation products (XLFDP's) it was shown that all
patients with DIC, confirmed by laboratory evidence of consumption coagulopathy and
clinical conditions characteristically associated with DIC, had levels ofXLFDP's above
the upper limit of 200 ng/ml that was observed in healthy controls." However, others
detected some overlap in XLFDP levels between patients with DIC and normal healthy
controls." Similar observations were reponed using an EIA specific for X-oligomers
(the earliest fmgment released from cross-linked fibrin): patients with DlC had elevated
levels of X-oligomers and showed no overlap with the levels of X-oligomers in normal
healthy controls.'l.l 60 In these studies. the clinical diagnosis of DIC was confirmed by
a low fibrinogen, elevated FDP in serum by latex assay, low platelet count (<10'/ml)
and a positive ethanol geh:rion test.
Using the D-dimer assay it was shown that all patients with DIC had elevated levels
of D-dimer (>200 ng/ml) and no overlap was shown with those in normal healthy
controls." However. I 0% of the hospitalized control group had elevated D-dimer
levels.87 Plasma of patient,; was examined one week before the onset of DIC (scored
according to Maegawa·s criteria of D!C) by D-dimer EIA. 159 Of the used laboratory
parameters (serum FDP. D-dimer. prothrombin time, fibrinogen and platelet count),
plasma D-dimer was the only parameter that was significantly elevated (no p value is
given) before the onset of DIC. while the other coagulation parameters became
significantly (p value unknown) abnormal only after the onset of DIC. This suggests
the possibility 10 identify an early state of DlC by D-dimer measurements in which
treatment may be started. ~nd requires further confirmation.
Plasmas of patients with DlC were also assayed by EIA's specific for FbDP's, FgDP's
or TDP' s. J9.26.30.l22.1S6-JS9
Patients with a high grade DIC had increased levels of FhDP's, TOP's and FgDP's as
39
compared to the levels in healthy controls.122 In these patients, the observed level of
FbDP was higher than the level of FgDP, suggesting that the process secondary
fibrinolysis is dominant and is accompanied by fibrinogenolysis.
Monitoring time dependant change of FbDP and TDP levels in plasma of patients with
DIC showed the possibility to detect the onset of primary fibrinolysis." In that study,
the FbDP and TDP levels were approximately equal during the first four days,
suggesting that nearly all degradation products are derived from fibrin. During the next
four days the levels of TDP exceeded the Fbdp levels indicating the occurrence of
fibrinogenolysis. The justification of this conclusion may follow from another study
where a good correlation (r=0.98) was observed between the values obtained by adding
the levels of FbDP and FgOP, and the actually measured TDP levels in patients with
DIC156
The possibility to use TDP. FbOP and FgOP levels, as assessed by EIA's, for the
monitoring of anticoagulant therapy (heparin and coumarin) and follow-up in patients
with chronic DIC was demonstrated in another study.'" In that study, the diagnosis of
DIC was based on the clinical picture and abnormalities in the following laboratory
parameters: platelet count. (pro)thrombin time. factor V, fibrinogen (Clauss), FM,
antithrombin-Ill, a,-antiplasmin and FOP in serum (Thrombo-Wellcotest). The levels
of FbDP and FgOP changed concomitantly, FbDP being higher than FgDP. This
indicates that both fibrinolysis and fibrinogenolysis occur. It was concluded that, of all
used laboratory parameters, soluble fibrin, TDP, FbDP and FgDP were the most
sensitive markers for acuve DIC and the effect of anticoagulant treatment. The
generally accepted view that an increased level of serum FOP's is required to permit
the diagnosis of DIC '2•36
·'·,. has to be revised.
Application ofthe EIA's showed that the secondary fibrinolysis in DIC is accompanied
by fibrinogenolysis. More studies are needed to determine the role of fibrin(ogen)
measurements by EIA's for the diagnosis of DIC. The EIA's for fibrin(ogen)
degradation products have not been applied to a population suspected of having OIC
to determine the sensitivity and specificity of these assays for the diagnosis of D!C.
Further studies are needed to determine these test characteristics and the predictive
40
values of the assays.
The recognition of the activity of each the two opposing processes coagulation and
fibrinolysis is important since it may have implications for a subsequent therapeutical
approach. Changes of plasma levels of fibrin(ogen) degradation products and soluble
fibrin, as detected by EIA's, were longitudinally measured in a patient with acute
promyelocytic leukemia. a condition known to be associated with DIC." The levels of
several fibrin degradation products, such as of D-dimer, FbDP's and X-oligomers
followed the same pattern in time. Surprisingly, an increase of these levels coincided
with decreased levels of soluble fibrin and vice versa. This suggest that valuable
information on the status of the haemostatic balance can be obtained by using one of
the EIA ·s for FbDP. D-dimer or X-oligomers (as markers of fibrinolysis) in tandem
with the soluble fibrin ElA (as a marker of coagulation). This should be confirmed in
other studies.
Coronary artery disease (CAD) and thrombolvtic therapy
It is well known that thrombotic occlusion of a coronary artery is
frequently associated with acute myocardial infarction (AMl). 161•162 Early recanalization
of these arteries in selected patients with AMI by fibrinolytic agents reduces short and
long term mortality and improves left ventricular function. 163-165 The clinical use of
thrombolytic agents is rapidly growing and a number of treatment strategies has been
used. 166 The success of this therapy is probably based on the induction of an extensive
fibrinolytic state. The recognition of an increased fibrinogenolytic state is important
since it is associated with severe bleeding complications. 167 But unfortunately, this
cannot be detected by the standard haemostatic screening tests like the PT, the APTT
or the bleeding time.
The quantitative plasma assays for fibrin(ogen) degradation products were applied to
patients with AMI in order to assess the intensity of fibrin(ogen)olysis induced by
thrombolytic therapy and to determine the role of fibrin(ogen) degradation products as
markers of this therapy. Increased levels of cross-linked fibrin degradation products
(XLFDP's), detected by EIA, have been detected in patients with AM!.' 63 Elevations
41
of plasma levels of XLFDP's above 1000 ng/ml occurred within one hour in all patients
receiving tissue-type plasminogen activator (t-PA) and in 83% of the patients treated
with streptokinase (SK).169 Seven hours after treatment, the XLFDP levels were
significantly elevated (p<0.01) in the t-PA treated group when compared to the
(declining) values that were observed in the SK treated group. This indicates that the
persistence of lysis of fibrin, as assessed by the ElA for XLFDP' s, is of longer duration
by administration of t-PA than following SK. The response of XLFDP to SK was
similar in the groups of patients with and without reperfusion, although in late samples
(> 7 hours) the elevations of XLFDP tended to be greater in the reperfusion group. The
difference. however. was statistically not significant.
Another study showed that plasma XLFDP levels increased 6 hours after t-PA infusion
as compared to pre-treatment values and no significant difference was observed
between the groups that received a dosages of 150 mg and 100 mg respectively. This
could indicate that the extents of lysis of fibrin are comparable in both groups. 170 In the
same study. the extent of fibrinogenolysis was assessed by using an EIA for BB1-42,
a peptide that is released when fibrinogen is degraded to fragment X.171 In contrast to
the levels of XLFDP's, the levels of plasma BBI-42 levels were significantly elevated
(p<O.OI) in the group receiving 150 mg t-PA as compared to values in the group given
100 mg. This indicates a more intense fibrinogenolysis following administration of the
higher dose. It might be concluded that increasing the dose of t-PA from !00 mg to
ISO mg results in an enhanced fibrinogenolysis, but has no influence on the extent of
secondary fibrinolysis. This finding was confirmed in another study 66 following n-PA
infusion in normal healthy volunteers. and may be used to define dose regimen since
it was demonstrated that a state of strong fibrinogenolysis is associated with bleeding
complications. 167 The study 66 demonstrates, that an increase of fibrin degradation
products. as assessed by the EIA for FbDP, after administration of n-PA is no evidence
for lysis of the thrombus, but merely for lysis of a body-pool of fibrin.
Systemic effects of thrombolytic therapy were also assessed by measuring plasma D
dimer levels by EIA. D-dimer levels were in the upper limit of the normal range (0.4
ug/ml) at the time of diagnosis for AMI before treatment and increased significantly
42
(p<O.OOI) 10 2.7ug/ml during rt-PA infusion. 172 However. elevated of plasma D-dimer
(of 980 ng/ml) were also observed in normal volunteers after t-PA infusion 173,
suggesting that another fibrin pool is lysed in addition 10 the coronary thrombus (e.g.
circulating soluble cross-linked fibrin polymers). As a consequence, an increase of D
dimer levels following t-PA administration is no evidence for lysis of the thrombus.
This view is supported by the observation that there is a discrepancy between
thrombolysis (i.e. reperfusion) and the elevation of plasma D-dimer levels."'
XLFDP levels were measured by EIA in patients with AMI presenting with and without
complications. 175 The complications were ventricular tachycardia or fibrillation requiring
electrical countershock. mural thrombus detected by echocardiography, severe
congestive heart failure defined as Class IV by M!RU criteria or death within 72 hours.
The used cut-off value for test positivity was 300 ng/ml. The sensitivity and specificity
of the EIA for the detection of complications in patients presenting early after onset of
symptoms (<8 hours) were 80% and 81% respectively. The prevalence of the
complications was 17% and the positive and negative predictive values were 47% and
95% respectively. In patients presenting late after (>8 hours) onset of symptoms the
sensitivity, specificity. positive and negative predictive values were 63%, 67%, 56%
and 73% respectively, at a prevalence of 40%. It can be concluded that normal levels
of XLFDP alone in patiems with AMI can not exclude the presence of complications.
No correlation was obser;ed between the size of the infarction and the levels of
XLFDP. The normal levels of XLFDP found in patients with uncomplicated AMI's
after early onset of symptoms, may be indicative for impaired fibrinolysis, as has been
suggested before.'"·m It was demonstrated that the plasma levels of XLFDP's in man
20 days (mean) after discharge from hospital following a myocardial infarction (MI)
were higher (p<O.OOJ) than the XLFDP levels in the age matched controls selected
from a general practice list, suggesting an increased thrombogenesis. fibrinolysis or
both in persons who suffered from Ml. 179
Using an EIA for FbDP. it was shown that patients with unstable angina had elevated
plasma levels of FbDP (]YO.OOI) when compared to norrnal controls or patients with
acute AMI.'" The FbDP le·;els in patients with stable angina was only slightly elevated
43
when compared to control levels (0.01 <p<0,02), but no difference was observed with
the levels in patients with uncomplicated AMI (p<O.OO!). The level of FbDP's
remained high in patients with recurrent episodes of chest pain, but declined in patients
who had only one episode of chest pain. It was suggested that significantly increased
FbDP levels in patients with unstable angina may be indicative for an ongoing
thrombolytic process, while the absence of elevated FbDP levels in patients with AMI
may suggest that clot dissolution is not taking place. However, this could not be
confirmed in another study by measuring plasma D-dimer levels by EIA. 181
Both patients with AMI and patients with unstable angina had moderate elevations of
plasma D-dimer as compared to patients with stable angina181 The sum of the FbDP
and FgDP levels. as measured by EIA, correlated (r=0.98) with the level of TDP in
patients with AMI following rt-PA treatment."' Normal values for FgDP and TDP in
plasma are below 0.25 ug!ml.''-''·' 82 A" rise of FgDP and TDP was observed in patients
with AMI two hours after infusion of SK.''-'6·182 The amount of FgDP was slightly less
than the amount of TDP. suggesting that nearly all circulating degradation products
were derived from fibrinogen."·''' The clearance of FgDP paralleled that of TDP and
a half-life time in the circulation of 5 hours was found. 182
T-PA infusion in patients with AMI resulted in elevated levels of FbDP, as detected by
EIA (mean: 2.7 ug/ml).'" As mentioned, increased levels of FbDP were also found in
normal healthy volunteers following rr-PA infusion (mean:0.8 ug/m\).66 In that study,
no significant differences in FbDP levels were found between volunteers who received
a rt-PA dose of 0.25mg/kg or 50mg/kg, suggesting that the extent of lysis is
comparable in the two groups. It was concluded that an increase of FbDP' s is not a
specific marker of intracoronary lysis. but also reflects systemic lysis of fibrin. This
view is supported by another study.""
In summary, application of the EIA for XLFDP's demonstrated that the lysis of fibrin
induced by administration of t-PA persists longer than by SK. The occurrence of
fibrinogenolysis. as assessed by an EIA for FgDP. has been demonstrated during rt-PA
and SK therapy and possibly contributes to an increased bleeding risk.167 Detection of
cross-linked degradation products by EIA can not be used to predict recanalization in
44
patients with AMI and the levels of XLFDP do not correlate to the size of an
infarction. Elevated plasma levels of fibrin degradation products are not specific
markers for the extent of fibrinolysis induced by thrombolytic therapy. The increase of
FbDP levels during thrombolytic therapy is probably caused by lysis of an other body
pool of fibrin outside the thrombus. Further studies are needed to determine the origin
of the fibrin pool that is lysed in addition to the coronary thrombus.
Malignancies
It has been shown that the coagulation system is activated in case of leukemia or solid
malignant tumors. 147'185
·1$
Tumor associated fibrinolysis is supposed to play a role in the local invasion or
metastatic dissemination of certain malignancies. Some studies demonstrated increased
levels of plasminogen actintors in the blood of patients with carcinomas which may
contribute to an enhanced fibrinolysis.' 85·19o.192 However. others reported a reduced
fibrinolytic activity in cancer patients. 193·194
The new assays were applied to measure the level of fibrin( ogen) degradation products
in patients with various malignant tumors.
With EIA"s for FgDP and D-dimer it was shown that patients with malignancies had
significantly higher plasma levels of FgDP (p<O.O!) and D-dimer (p < 0.05) as
compared to the levels in age-sex matched healthy controls. 185 This suggests that
fibrinolysis is accompanied by fibrinogenolysis in patients with malignancies.
In tissue specimen of patients with ovarian cancer (Figo stage Ill and IV) or breast
cancer (T,-TJN0-N2) elevated levels of D-dimer. as assessed by EIA. were found
relative to D-dimer levels' in tissue specimen of benign controls.195 It was shown that
the D-dimer levels in ovarium cancer tissue were higher than in tissue of patients with
breast cancer. This was confirmed in another study.196
Using an EIA for D-Dimer measurements in plasma it was shown that patients with
ovarian cancer or cervical carcinoma had significantly (p < 0.01) elevated levels of
plasma D-dimer when compared to D-dimer levels in norrnal healthy controls.'"
These findings suggest that fibrinolysis is a common feature in patients with
45
malignancies. With an ElA for the assessment of TOP's. it was shown that 74% of
patients with ovarium cancer treated by chemotherapy had increased levels of TDP' s
as compared to normal values (<110 ng/ml). 19 In a majority of these patients (no
percentage is mentioned) a decrease of the tumor mass (serially assessed by
laparoscopy, CT scanning 0r second look operations) was accompanied by a decline in
plasma TDP levels. Since it is possible to detect low levels of TDP (i.e. patients in
partial remission) with this EIA, it may be particularly valuable in combination with
the tumor marker CA 125. The latter has been shown to lack sensitivity in detecting
small tumor masses, which are observed by second look operations.198 More studies are
needed to confirm these findings.
By E!A's, elevated plasma levels of fibrin(ogen) degradation products were also found
in patients with carcinoma of the liver (FbDP,FgDP and TOP 122), lung cancer (0-
Dimer 200) and acute promyelocytic or myeloid leukemia (D-Dimer 87
•199
; XDP 157•
FbDP, FgDP, TDP and X-oligomer 199).
The studies mentioned above. using the new plasma assays, demonstrate that the levels
of fibrin(ogen) degradation products are elevated in patients with malignancies, which
may indicate that the fibrinolytic system remains active in these patients.
Conclusions
In this paper we discussed the possible clinical utility of assays for determination of
degradation products of fibrin and fibrinogen. We pointed out that assays based on
polyclonal antibodies and performed in serum are unreliable and therefore their clinical
use is limited. Until recently it was not possible to determine the extent of primary
activation of the fibrinolytic system specifically. The new assays are performed on
plasma and make it possible to quantify the extent of fibrin(ogen)olysis reliably.
Furthermore they allow discrimination between the degradation products of fibrin and
of fibrinogen, since a specific assay has been developed for separate and specific
determination of fibrinogen degradation products.
Fibrinolysis appears to be associated with (low grades of) fibrinogenolysis. The process
of secondary fibrinolysis, as derived from FbDP and D-dimer levels, appears to
46
predominate in deep venous thrombosis, pulmonary embolism, sickle cell disease,
myocardial infarction and unstable angina. It was shown that thrombolytic therapy
results in a pronounced fibrinogenolysis (primary fibrinolysis), as assessed by the levels
of FgDP. The occurrence of an extensive fibrinogenolysis was also observed during
orthotopic liver transplantation. Several diseases, such as DIC, malignancies, liver
disease and renal diseases, exhibit both fibrinogenolysis and fibrinolysis secondary to
an activated state of coagulation.
The EIA for plasma D-dimer can be used to exclude the diagnosis in patients clinically
suspected for DVT and ascending venography is not required. The semiquantitative
latex assays for D-dimer can not be used for this purpose. Plasma FbDP levels, as
detected by EIA, reflect decrease in thrombus size and may be used to monitor efficacy
of heparin treatment. Low initial values of FbDP are predictive of poor thrombus
dissolution in patients with DVT following SK or UK therapy. Elevated plasma levels
of cross-linked fibrin degradation products have been measured during thrombolytic
therapy of AMI. They are derived mainly from extracoronary sites and are non
predictive of recanalization following thrombolytic therapy.
Plasma Fibrin degradation products levels are related to severity of liver cirrhosis and
renal disease, but can not be used as markers of renal function. The high sensitivity of
the EIA 's made it possible to detect increased levels of fibrin degradation products
during steady state in patients with sickle cell disease. The observed increased
fibrin(ogen) levels during normal and complicated pregnancy suggest that fibrinolysis
is not necessarily depressed. The fibrinolytic system remains active in patients with
malignancies. as demonstrated by elevated levels of D-dimers and FbDP's.
The fo!lowing guidelines for future studies are suggested. As mentioned in the
introduction, a disturbance in the haemostatic balance will presumably be reflected in
the products of both coagulation and fibrinolysis. In this article we focussed on
fibrin(ogen)olysis only. Detection of fibrin(ogen) derivatives by the monoclonal
antibody based assays provides valuable information on haemostasis in several diseases.
However, it is conceivable that more information can be obtained when the products
of the two opposing processes coagulation and fibrinolysis are simultaneously
47
measured. This possibility is available. Soluble fibrin, the product of an activated
coagulation system, can also be assessed by EIA in plasma2"
25 In the future, studies
may be performed using a combination of an EIA for a coagulation product (i.e. soluble
fibrin) and an EIA for a fibrinolysis product (i.e. FbDP or D-dimer).
REFERENCES l. Donati MB. Assays for fibrinogen/fibrin degradation products in biological fluids: some
methodological aspects Thromb Diath Haemorr 1975; 34: 652 - 662.
2. Ruckley CV, Das PC, Leitch AG,Donaldson, AA, Copland WA, Redpath AT, Scott P,
Cash JD. Serum fibrin/fibrinogen degradation products associated with postoperative
pulmonary embolus and venous thrombosis. Brit Med J 1970; 4: 395-398
3. Hedner U, Nilsson I M. Clinical experience with determination of fibrinogen degradation
products. Acta Med Scand 1971: 189: 471-477.
4. Wood EH. Prentice CRM, McNicol GP. Association of fibrinogen- fibrin related antigen
(F.R. antigen) with postoperative deep vein thrombosis and systemic complications.
Lancet 1972: i: 166- 169.
5. Gallus AS, Hirsch J, Gent M. Relevance of preoperative and postoperative blood tests
to postoperative leg vein Thrombosis. Lancet 1973; ii: 805 - 809.
6. Gurewich V, Hume M, Patrick M. The laboratory diagnosis of venous thromboembolic
disease by measurement of fibrinogen/fibrin degradation products and fibrin monomer.
Chest 1973; 64: 585-590.
7. Cooke ED, Gordon YB, Bowcock SA, Sola CM, Pilcher MF, Chard T, Ibbotson RM,
Ainsworth ME. Serum fibrin(ogen) degradation products in diagnosis of deep vein
thrombosis and pulmonary embolism after hip surgery. Lancet 1975; ii: 51-54.
8. Clayron JK, Anderson JA, McNicol GP. Preopemtive prediction of postoperative deep
vein thrombosis. Brit Med J 1976: 2: 910-912.
9. Leslie J. Langler D, Serjeant G, Serjeant B, Desai P, Gordon Y. Coagulation changes
during the steady state in homozygous sickle-cell disease in Jamaica. Br J Haematol
1975; 30: 159-166.
10. Gaffney PJ, Perry MJ. "Giant" fibrin fragments and thrombosis. Thromb Haemostas 1985
; 54: 931 (abstract).
11. Gaffney PJ, Creighton LC, Harris R, Perry MJ. Monoclonal antibodies (MABS) to
48
crosslinked fibrin fragments. Their characterization and potential clinical use. In: Miiller
Berghaus G, Scheefers-Borchel U, Selmayr E, Henschen A, eds. Fibrinogen and its
derivatives. Biochemistry, physiology and pathophysiology. AmSterdam. Excerpta
Medica, 1986; 273-284
12. Bick RL. Disseminated intravascular coagulation: a clinical/laboratory study of 48
patients. Ann NY Acad Sci 1981; 370: 843.
13. Kies MS, Posch JJ, Giolma JP, Rubin RM. Hemostatic function in cancer patients.
Cancer !980; 46: 831.
14. Tibbut DA, Chesterman CN, Allington MJ, Williams EW, Faulkner T. Measurement of
fibrinogen-fibrin related antigen in serum as aid to diagnosis of deep vein thrombosis in
outpatients. Brit Med J 1975; 43: 493.
15. Gaffney PJ, Chesterman GN, Allington MJ. Plasma fibrinogen and its fragments during
streptokinase treatment. Br J Haematol 1974; 26: 285-93
16. Francis CW, Marder VJ, Barlow GH. Plasmic degradation of cross linked fibrin. Biochim
Biophys Acta 1973; 295: 308-313.
17. Doolittle RF. Structural aspects of the fibrinogen to fibrin conversion. Adv Protein Chern
1973; 27: l-109.
18. Brass EP, Forman WB, Edwards RV, Lindan 0. Fibrin formation: the role of the
fibrinogen-fibrin monomer complex. Thromb Haemostas 1976; 36: 36-48.
19. Koopman J, Haverkate F, Koppen PW, Nieuwenhuizen W, Brommer EJP, van der Werf
WGC. New enzy~e immunoassay of fibrin-fibrinogen degradation products in plasma
using a monoclonal antibody. J Lab Clin Med 1987; 109: 75-84.
20. Doolittle RF. Fibrinogen and fibrin. Sci Am !981; 245: 92-101.
21. Hoegee-de Nobel E, Voskuilen M, Briet E, Brommer EJP, Nieuwenhuizen W. A
monoclonal antibody-based quantitative enzyme immunoassay for the determination of
plasma fibrinogen concentrations. Thromb Haemostas 1988; 60: 415-418.
22. Koppen PW, Huijsmans CMG, Nieuwenhuizen W. A monoclonal antibody, specific for
human fibrinogen, fibi.nopeptide A-containing fragments and not reacting with free
fibrinopeptide A. Blood 1985; 66: 503-507.
23. Mi.iller-Berghaus G, Scheefers-Borchel U, Fuhge P, Eberle R, Heimburger N. Detection
of fibrin in plasma by a monoclonal antibody against the aminoterminus of the alpha
chain of fibrin. Scand J Lab Invest !985; 45: 145-151.
49
24. Scheefers-Borchel U. Muller-Berghaus G, Fuhge P, Eberle R, Heimburger N.
Discrimination betwee~ fibrin and fibrinogen by a monoclonal antibody against a
synthetic peptide. Proc Nat! Acad Sci USA 1985; 82: 7091-7095.
25. Scheefers-Borchel U, Mliller-Berghaus G, Fuhge P, Heimburger N. Discrimination
between fibrin and fibrinogen by a monoclonal antibody against a synthetic hexapeptide
presenting the amino-tenninus of the alpha-chain of fibrin. In: Lane DA, ed. Fibrinogen,
fibrin formation and fibrinolysis. Berlin, Walter de Gruyter, 1986; 253-260.
26. Koppen PW, Kuipers W, Hoegee-de Nobel E, Brommer EJP, Koopman J,
Nieuwenhuizen W. A quantitative enzyme immunoassay for primary fibrinogenolysis
products in plasma. Thromb Haemostas 1987; 57: 25-28.
27. Koppen PW, Koopman J, Haverkate F, Nieuwenhuizen W. Production and
characterization of a monoclonal antibody reactive with a specific neoantigenic
determinant (comprising BB 54-118) in degradation products of fibrin and of fibrinogen.
Blood 1986; 68: 437-441.
28. Elms MJ, Bunce JH, Bundesen PG, Rylan DB, Webber AJ, Masci PP, Whitaker AN.
Measurement of crosslinked fibrin degradation products - An immunoassay using
monoclonal antibodies. Thromb Haemostas 1983; 50: 591-594.
29. Rylatt DB, Blake AS, Glottis LE, Massingham DA, Fletcher WA, Masci PP, Whitaker
AM, Elms M, Bunce J, Webber A, Wyatt D, Bundesen PG. An immunoassay for human
D-dimer using monoclonal antibodies. Thromb Res 1983: 31: 767-778.
30. Koppen PW. Hoegee-de Nobel E, Nieuwenhuizen W. A monoclonal antibody-based
enzyme immunoassay for fibrin degradation products in plasma. Thromb Haemostas
1988: 59: 3!0-3!5.
31. Graeff H, Hafter R, von Hugo R. On soluble fibrinogen-fibrin complexes. Thromb Res
1979: 16: 575-576.
32. Hermans J, McDonagh J. Fibrin: structure and interactions. Sern Thromb Haemostas
1982; 8: 11-24.
33. Holm B, Nilsen DWT, Kierulf P, Godal HG. Purification and characterization of 3
fibrinogens with different molecular weights obtained form normal human plasma.
Thromb Res !985; 37: 165-176.
34. Moses son MW, Galanakis DK, Finlayson JS. Comparison of human plasma fibrinogen
subtractions and early plasmic fibrinogen derivatives. J Bioi Chern 1974; 249: 4656-
50
4664.
35. Philips HM. Isolation of six homogeneous preparations of high molecular weight and low
molecular weight fibrinogen. Can J Biochem 198!; 59: 332-342.
36. Graeff H, Hafter R. Detection and relevance of crosslinked fibrin derivatives in blood.
Sem Thromb Haemostas !982; 8: 57-68.
37. Merskey C, Lalezazi P, Johnson AJ. A rapid simple sensitive method for measuring
fibrinolytic split products in human serum. Proc Soc Exp Biol Med NY 1969; 131: 871-
875.
38. Connaghan DG, Francis DW, Ryan DH, Marder VJ. Prevalence and clinical implications
of heparin-associated false positive tests for serum fibrin(ogen) degradation products. Am
J Clin Patho! 1986: 86: 304-10.
39. Haverkate F, Timan G, Nieuwenhuizen W. Anticlorring properties of fragments D from
human fibrinogen and fibrin. Eur J Clin Invest 1979; 9: 253-255.
40. Nieuwenhuizen W, Gravesen M. Anticoagulant and calcium-binding properties of high
molecular weight derivatives of human fibrinogen, produces by plasmin (fragments X).
Biochim Biophys Acta 1981; 668: 81-88.
41. Nieuwenhuizen W, Voskuilen M, Hermans J. Anticoagulant and calcium-binding
propenies of high molecular weight derivatives of human fibrinogen (plasmin fragments
Y). Biochim Biophys Acta 1982; 708: 3!3-316.
42. Marder YJ, Shulman NR. High molecular weight derivatives of human fibrinogen
produced by plasmin II. Mechanism of their anticoagulant activity. J Biol Chern 1969;
244: 2120-2124.
43. Gaffney PJ, Perry MJ. Unreliability of current serum fibrin degradation products (FDP)
assays. Thromb Haemostas 1985; 53: 30!-302.
44. Niewiaroski S, Stewan GJ, Marder Y J. Formation of highly ordered polymers fibrinogen
and fibrin degradation products. Biochim Biophys Acta 1970; 221: 326-341.
45. Nieuwenhuizen W. A double blind comparJ.tive study of six monoclonal antibody-based
plasma assays for fibrinogen derivatives. In: Lowe GDO, Douglas IT. Forbes CD,
Henschen A, eds. Fibrinogen 2. biochemistry. physiology and clinical relevance.
Amsterdam, Excerpta Medica !987; 18!-!86.
46. Mirshahi M, Soria J, Soria C, Perrot JY, Boucheix C. A latex immunoassay of
fibrin/fibrinogen degradation products in plasma using a monoclonal antibody. Thromb
51
Res 1986; 44: 715-728.
47. Buller HR, Ockelford PA, Hull RD. Strategies of diagnosis and screening of deep venous
thrombosis and pulmonary embolism.
Neth J Surg 1983; 35: i21-128.
48. Huisman MY, Buller HR, Ten Cate JW, Vreeken J. Serial impedance plethysmography
for suspected deep venous thrombosis in outpatients. N Eng! J Med 1986; 314: 823-828.
49. Elms MJ, Bunce !A, Bundesen BG, Rylatt DB, Webber AJ, Masci PP, Whitaker AN.
Rapid detection of cross-linked fibrin degradation products in plasma using monoclonal
antibody-coated particles. Am J Clin Path 1986; 85: 360-364.
50. Hillyard CJ, Blake AS, Wilson K, Rylatt DB, Miles S, Elms MJ, Barnes A, Bundesen
PG. Latex agglutination assay for D-dimer: Evaluation and application to the diagnosis
of thrombotic disease. Clin Chern 1987; 33: 1837-1840.
51. Heaton DC. Billings JD, Hickton CM. Assessment of D dimer assay for the diagnosis
of deep vein thrombosis. J Lab Clin Med 1987; 110: 588-591.
52. Bounameaux H, Schneider PA, Reber G, De Moerloose P, Krahenbuhl B. Measurement
of plasma D-dimer for Diagnosis of deep venous thrombosis. Am J Clin Path 1989; 91:
82-85.
53. Ott P, Asrrup L, Jensen RH, Nyeland B, Pedersen B. Assessment of D-Dimer in plasma:
Diagnostic value in suspected deep venous thrombosis of the leg. Acta Med Scand 1988;
224: 263-267.
54. Hochstein BE, Johns AS, Ockelford PA. D dimers (dimenest): a screening for the
diagnosis of venous thrombosis. Thromb Haemostas 1989; 62: 124 (abstract).
55. Whitaker AN, Elms IVu, Masci PP, Bundesen PG, Rylal! DB, Webber AJ, Bunce !H.
Measurement of cross linked fibrin derivatives in plasma: an immunoassay using
monoclonal antibodies. J Clin Pathol 1984: 37: 882-887.
56. Rowbotham BJ, Carroll P, Whitaker AN, Bunce lA, Cobcroft RG, Elms MJ, Masci PP,
Bundesen PG, Rylan DB, Webber AJ. Measurement of crosslinked fibrin derivatives- use
in the diagnosis of venous thrombosis. Thromb haemostas 1987; 57: 59-61.
57. Van Bergen PFMM, Knot EAR, Jonker JJC. De Boer AC, De Maat MPM. Is quantitative
determination of fibrin(ogen) degradation products and thrombin-antithrombin lll
complexes usefull to diagnose deep venous thrombosis in outpatients? Thromb
Haemostas 1989; 62(4):1043-1045.
52
58. Mirshahi M, Soria C, Soria J, Mirshahi M, Faivre R, Kieffer Y, Bassand JF, Toulemonde
F, Caen J. Changes in plasma fibrin degradation products as a marker of thrombus
evolution in patients with deep venous thrombosis. Thromb Res 1988; 51: 295-302.
59. Faivre R, Mirshahi M, Ducellier D, Soria C, Soria J, Mirshahi M, Kieffer Y, Bassand JF,
Caen J, Maurat JP. Evolution of plasma specific fibrin degradation products during
thrombolytic therapy in patients with thrombo-embolism. Thromb Res 1989; 50; 583-
589.
60. Bridey F, Phlipotteau C, Dreyfus M, Simmoneau G. Plasma D-dimer and pulmonary
embolism. Lancet 1989: 791. (letter).
61. Gaffney PJ ,Creighton U, Callus M, Thorpe R. Monoclonal antibodies to crosslinked
fibrin degradation products (XL-FDP). II Evaluation in a variety of clinical conditions.
Br J Haematol 1988; 68: 91-96
62. Bounameaux H, Slosman D, de Moerlose Ph, Reber G. Diagnostic value of plasma D
dimer in suspected pulmonary embolism. Lancet 1988; ii: 628-29.
63. Goldhaber SZ, Vaughan DE , Tumeh SS, Loscalzo J. Utility of cross-linked fibrin
degradation products in the diagnosis of pulmonary embolism. Am Hean J 1988; 116:
505-508
64. Speiser W. Leitha T, Dudczak R, Lechner K. lancet 1989: 792. (letter).
65. Faivre R, Ducel!ier MD, Neuhart E. Mirshahi M, Soria C, Caen J, Bassand JP, Maurat
JP. How long must be given thrombolytic therapy for massive venous thromboembolism?
Thromb Haemostas 1989: 62: 546 (absrract).
66. Seifried S, Tanswell P, Rijken DC, Kluft C, Hoegee E, Nieuwenhuizen W. Fibrin
degradation products are not specific markers for thrombolysis in myocardial infarction.
Lancet 1987: 333-334.
67. Amir-Ahmadi H, McGray RS, Martin F, Mitch W, Kantrowitz P, Zamcheck H.
Reassessment of massive upper gastrointestinal hemorrhage on the wards of the BostOn
City Hospital. Surg Clin North Am 1969: 49: 715.
68. Ratnoff OD, Patek AJ Jr. The natural hisrory of Laennec's cirrhosis of the liver: an
analysis of 386 cases. Medicine 1942; 21: 207-262.
69. Schlichting P, Christensen E, Fauerholdt L, Poulsen H, Juhl E, Tytstrup & CSL. Main
causes of death in cirrhosis of the liver. Scand J Gastroenterol 1983; 18: 881-888.
70. Benaglia E, Belmonte P, Vertolli U, Azzurro M, Martines D. Bleeding in cirrhotic
53
patients: A precipitating factor due tO intravascular coagulation or to hepatic failure?
Haemostasis 1983; 13: 328-334.
71. Bloom AL. lnrravasculcr coagulation and the liver. Br J Haematol 1975; 30: 79-81.
72. Straub PW. Diffuse intravascular coagulation in liver disease? Sem Thromb Haemostas
1977; 4: 29-39.
73. VanDeWater L, Carr JM, Aronson D, Me Donagh J. Analysis on elevated fibrin(ogen)
degradation product levels in patients with liver disease. Blood 1986; 67: 1468-1473.
74. Verstraete M, Vennylen J, Collen D. Intravascular coagulation in liver disease. Annu Rev
Med 1974; 25: 447-455.
75. Marongiu F, Mameli G. Acca M R, Mamusa M, Mulas G, Balestrieri A. Low grade DIC
in liver cirrhosis: facr. or fiction? Thromb Haemostas 1988; 59: 344. (lener).
76. Fiori G, Mombelli G, Haeberli A, Straub PW. Evidence against DIC in liver cirrhosis.
Thromb Haemostas 1987: 58: 77 (abstract).
77. Oka F, Tanaka K. Intravascular coagulation in autopsy cases with liver disease. Thromb
Haemostas 1979; 42: 564-570.
78. Tytgat G, Collen D, De Yreker R, Verstraete M. Investigations on the fibrinolytic system
in liver cirrhosis. Acta Haematol 1968; 40: 265-274.
79. Knot EAR, Drijfhout HR, Ten Cate JW, DeJongE, Iburg AHC, Kahle LH, Grijm R. a,
Plasmin inhibitor mechanism in patients with liver cirrhosis. J Lab Clin Med 1985; 105:
353-358.
80. Leebeek FWG, Kluft C, Knot EAR, De MaatMPM. Inappropriate inhibition at the
plasmin level may contribute to enhanced fibrinolysis in advanced liver cirrhosis.
Fibrinolysis 1988; 2: 75-78.
81. Hersch S, Kunelis T, Francis RB Jr. Pathogenesis of accelerated fibrinolysis in liver
cirrhosis: a critical role for tissue plasminogen activator inhibitor. Blood 1987; 69: 1315-
1319.
82. Boks A, Brommer EJP, Schalm SW, Van Vliet HHDM. Haemostasis and fibrinolysis in
severe liver failure and their relation to hemorrhage. Hepatology 1986; 6: 79-86.
83. Francis RB JR, Feistein Dl. Clinical significance of accelerated fibrinolysis in liver
disease. Haemosrasis 1984; 14: 460-465
84. Nilsson T, Wallen P, Mellbring G.ln vivo metabolism of human tissue-type plasminogen
activator. Scand J Haematol 1984; 33: 49-53.
54
85. De Maar MPM, Nieuwenhuizen W, Knot EAR. Discrepancies between levels of intact
and functional fibrinogen in patients with liver cirrhosis. Thromb Haemostas 1989; 62:
79 (abstract).
86. Palareti G, Legnani M, Maccaferri M, De Rosa V, Fonunato G, Grauso F, Sarna C,
Bellusci R, Franceschelli N, Coccheri S. Hyperfibrinolysis and D!C during onhotopic
liver transplantation (OLT). Effect of antithrombin III (ATIII) concentrates
administration. Thromb Haemostas 1989; 62: 179 (abstract).
87. Wilde JT, Kitchen S, Kinsey S, Greaves M, Preston PE. Plasma D-dimer levels and their
relation to serum fibrinogen/fibrin degradation products in hypercoagulable states. Br J
Haematol 1989; 71: 65-70.
88. Kurokawa I, Kondoh M. Determination of FDP D-dimer in human blood by particle
counting immunoassay and its clinical significance. Thromb Haemostas 1989; 62: 234
(abstract).
89. Lewis JH. Bontempo FA, Cornell FW, Kiss JE, Larson P, Ragni MV, Rice EO, Spero
JA, Starzl TE. Blood use in liver transplantation. Transfusion 1987; 27: 222- 225.
90. Goldsmith MF. Liver transplants: Big business in blood. JAMA 1983; 250: 2904.
91. Groth CG. Changes in coagulation. In Starzl TE. Putman CW, eds. Experience in
hepatic transplantation. Philadelphia: WB Saunders, 1969: 159-175.
92. Lewis JH, Bontempo FA, Kang YG, Spero JA, Ragni MV, Starzl TE. In :Winter PM,
Kang YG, eds. Hepatic transplantation. New York: Praeger Publishers 1986: 142-150.
93. Palaretto, De Rosa Y. Fortunato G, Grauso F. Legnani C, Maccaferri M, Poggi M,
Bianchini B, Bellusci R, Franceschelli N, Coccheri S. Control of haemostasis during
onhotopic liver transplantation. Fibrinolysis 1988; 2: 61-66.
94. BOhmig HJ. The coagulation disorder of orthotopic hepatic transplantation. Sem Thromb
Hemostas 1977: 4: 57-82
95. Von Kaulla KN, Kayne H, Von Kaulla E, Marchioro TL, Starzl TE. Changes in blood
coagulation before and after hepatectomy or transplantation in dogs and man. Arch Surg
1966; 92: 71-79
96. Pone RJ, Bontempo FA, Knot EAR, Lewis JH, Kang YG. Systemic effects of tissue
plasminogen activator associated fibrinolysis and the relation with thrombin generation
in onhotopic liver transplantation. Transplantation 1989; 47: 978-984.
97. Le Querrec A, Derlon A, Gobin J, Deshayes JP, Segol P, Bricard H, Thomas M.
55
Disseminated intravascular coagulation and hyperfibrinolysis during onhotopic liver
transplantation. Thromb Haemostas 1989; 62: 179 (abstract).
98. Dzik WH. Arkin CF. Jenkins RL, Stump DC. Fibrinolysis during liver transplantation in
humans: role of tissue-type plasminogen activator. Blood 1988; 71: 1090-1095.
99. Harper PL, Luddington RJ, Jennings l, Reardon D, Seaman MJ, Carrell RW, Kllink JR,
Smith M, RoUes K, Calne R. Coagulation changes following hepatic revascularization
during liver transplantation. Transplantation 1989; 48: 603-607.
100. Kratzer MAA, Dieterich HJ, Denecke H, Knedel M Mechanism of activation of
coagulation during orthotopic human liver transplantation. Thromb Haemostas 1989; 62:
180 (abstract).
101. Knot EAR, Porte RJ, Terpstra OT, Coagulation and fibrinolysis in the first human
auxiliary partial liver transplantation in Rotterdam. Fibrinolysis 1988; 2: 111-117.
102. Leichtman DA, Brewer GJ. Elevated plasma levels of fibrinopeptide A during sickle cell
anemia pain crisis -evidence for intrava.'\Cular coagulation. Am J Hematol 1978; 5: 183-
190.
103. Diggs L. Sickle cell crises. Am J Clin Path 1965: 44: 1-16.
104. Mattii R, Weinger R, Sise H. Coagulation, fibrinogen survival, and fibrin split products
in sickle cell disease. Blood 1973; 42: !004 (abstract).
105. Richardson SGN. Matthews KB, Stuart J, Geddes AM, Wilcox RM. Serial changes in
coagulation and viscosi;:y during sickle-cell crisis. Br J Haematol 1979; 41: 95.
106. Alkjaersig N, Fletcher A, Joist H, Chaplin H. Hemostatic alterations accompanying sickle
cell pain crisis. J Lab Clin Med 1976; 88: 440.
107. Devine DV, Kinney TR, Thomas PF, Rosse WF, Greenberg CS. Fragment D-Dimer
levels: An objective marker of vasa-occlusive crisis and other complications of sickle cell
disease. Blood 1986; 68: 317-319
108. Francis Jr. RB. Elevated fibrin D-Dimer fragment in sickle cell anemia: Evidence for
activation of coagulation during the steady state as well as in painful crisis. Haemostasis
1989: 19: 105-!ll.
109. McKay DG. Chronic intravascular coagulation in normal and preeclampsia. Contr
Nephrol !981: 25: 108-1!9.
110. Hustin J, Foidart JM, Lambotte R. Maternal vascular lesions in preeclampsia and
intrauterine growth retardation: light microscopy and immunofluorecence. Obstet Gynec
56
Surv 1985; 40: 145-149.
111. Terao T. Kobayashi T. The role of placental urokinase inhibitor in toxemia of pregnancy.
Bioi Res Pregnancy Perinarol 1983; 4: 145-151.
112. Gordon YB, Ratky SM, Baker LRI, Letchworth AT, Leighton PC, Chard T. Circulating
levels of fibrin/fibrinogen degradation fragment E measured by radioimmunoassay in
preeclampsia. Br J Obstet Gynecol 1976; 83: 287-291
113. Gibson B, Hunter D, Nearne PB, Kelron JG. ThrombocytOpenia in preeclampsia and
eclampsia. Sem Thromb Hemostas 1982; 8: 234-247
114. Bunnar J, McNicol GP, Douglas AS. Coagulation and fibrinolytic systems in pre
eclampsia and eclampsia. Br Med J 1975; 1: 12
115. Pritchard A, CunninghJ.m G, Masom MA: Coagulation changes in eclampsia: their
frequency and pathogenesis. Am J Obstet Gynecol 1974; 124: 855
116. Roberts JM, May JW. Consumption coagulopathy in severe pre-eclampsia. Obstet
Gynecol 1976; 48: 163
117. Trofatter KF, Howell ML, Greenberg CS, Hage ML. Use of the fibrin D-Dimer in
screening for coagulation abnormalities in preeclampsia. Obstet Gynecol 1989; 73: 435-
439.
118. Kobayashi T, Terao T. Preeclampsia as chronic disseminated intravascular coagulation.
Gynecol Obstet Invest 1987; 24: 170-178.
119. Rossi L, Vannini F, Felloni L, Banolai S, Lechinni L, Panicucci F: evaluation of D
Dimer with monoclonal antibodies and its clinical importance. Thromb Haemostas 1989;
62: 236 (abstract).
120. Woodhams BJ, Candotti G, Shaw R, Kernoff PBA. Changes in coagulation and
fibrinolysis during pr"!gnancy: evidence of activation of coagulation preceding
spontaneous abortion. Thromb Res 1989; 55; 99-107.
121. Ballegeer V, Mombaerts P. Declerck PJ, Spitz B, Van Assche FA, Collen D. Fibrinolytic
response to venous occlusion and fibrin fragment D-Dimer levels in normal and
complicated pregnancy. Thromb Haemostas 1987; 58: 1030-!032.
122. Rossi L, Felloni L, Ya•mini F, Lechinni L, Panicucci F. Evaluation of fibrinogen and
fibrin degradation products in physiological and pathological conditions using monoclonal
antibodies. Thromb Haemostas 1989; 62: 232 (abstract).
123. Ostlund E, Almqvist l, Blomback M, Bremme K, Wallin M. Activation of coagulation
57
and fibrinolysis during normal pregnancy. Thromb Haemostas 1989; 62: 236 (abstract).
124. Thorsen S. The inhibition of tissue plasminogen activator and urokinase induced
fibrinolysis by some natural proteinase inhibitOrs and by plasma and serum from normal
and pregnant subjects. Scand J Clin Lab Invest 1973; 31: 51-59.
125. Walker JE, Gow L, Cambell DM, Ogston D. The inhibition by plasma of urokinase and
tissue activator induced fibrinolysis in pregnancy and puerperium. Thromb Haemostas
1983; 49: 21-23.
126. Gore M, Eldon S, Trofatter KF, Soong SJ, Pizzo SV. Pregnancy induced changes in the
fibrinolytic balance: evidence for defective release of tissue plasminogen activator and
increased levels of fast acting tissue plasmnogen activator inhibitor. Am J Obstet
Gynecol 1987; 156: 674-680.
127. Morita T, Susuki Y, Churg J. Structure and development of glomerular crescent. Am J
Path 1973: 72: 349.
128. Kincaid-Smith P. Coagulation and renal disease. Kidney lnt 1972: 2: 183-190.
129. Koffler N. Paronetto F. Immunofluorescent localization of immunoglobulins, complement
and fibrinogen in human disease. J Clin Invest 1965; 45: 1665.
130. McDonald MK, Clarksrm AR, Davidson AM. The role of coagulation in renal disease.
In: Kincaid-Smith P, Mathew Th, Becker EL eds. Glomerulonephritis. John Wiley and
sons, New York I 973: 923.
131. McCluskey RT, Yassalli R, Gallo G. An immunofluorescent study of pathogenic
mechanism in glomerular disease. N Eng! J Med 1966; 274: 695.
132. Steihm ER, Trygstad CW. Split products of fibrin in human renal disease. Am J Med
1969; 46: 774.
133. Hedner U, Nillson IM. Renal diseases and fibrinogen degradation products. In:
Hamburger J, Crosnier J, Maxwell MH Eds. Advances in nephrology. Year Book
Medical Publishers, Chicago, Illinois 1973: 241.
134. Hedner U. Urinary fibrin/fibrinogen derivatives. Thromb Diath Haemorrh 1975; 34:693.
135. McNicol GP, Prentice C..RM, Briggs JD. Fibrinogen degradation products or FDP in renal
disease : estimation and significance of FDP in urine. Scand J Haematol 1971; 13: 329
136. Portemont G, Vermylen J, Donati MB. Urinary excretion of fibrinogen/fibrin related
antigen in glomerulonephritis. In: Kincaid-Smith P, Mathew Th, Becker EL eds.
Glomerulonephritis. John Wiley and sons, New York 1973: 829
58
137. Sakakibara K, Nagase M, Takada Y, Takada A. Relationship between urinary fibrinogen
degradation products and various types of chronic nephritis. Thromb Res 1987; 45: 403-
411.
138. Kamitsuji H, Matsunaga T, Taira K, Nakajima M, Whitworth JA, Kincaid-Smith P:
Urinary cross-linked fibrin degradation products in glomerular disease with crescents.
Clin Nephrol 1988; 29: 124-128.
139. Karnitsuji H, Whitworth JA, Dowling JP, Kincaid-Smith P. Urinary Crosslinked fibrin
degradation products in glomerular disease. Am J Kid Dis 1986; 7: 452-455.
140. Taira K, Matsunaga T, Kawahara S, Sakamoto S, Kamitsuji H. Fragments of urinary
fibrin/fibrinogen degradation products and cross-linked fibrin degradation products in
various renal diseases. Thromb Res 1989; 53; 367-377.
141. Gordge NMP, Faint RW, Rylance PB, Ireland H, Lane DA, Neild GH. Plasma D-Dimer:
A useful marker of fibrin breakdown in renal failure. Thromb Haemostas !989; 61: 522-
525.
!42. Clarkson AR, Monon JB, Cash JD. Urinary fibrin/fibrinogen degradation products after
renal homotransplantation. Lancet 1970; ii: 1220.
143. Siegal T, Seligsohn U, Aghai E, Madan M. Clinical and laboratory asspects of
disseminated intravascular coagulation (DIC): A study of 118 cases. Thromb Haemostas
1978; 39: 122-140.
144. Coleman RW, Robboi SJ, Mina JD. Disseminated inn-avascular coagulation (DIC): an
approach. Am Rev Med 1972; 52: 679-689.
!45. Thomas DP. Niewiarowski S, Myers AR, Block KJ, Colman RW. A comparative study
of four methods for detecting fibrinogen degradation products in patients with various
diseases. N Eng! J Med 1970; 238: 663-668.
146. Spero JA, Lewis JH, Ha.siba U. Disseminated intravascular coagulation: Findings in 346
patients. Thromb Haemostas 1980: 38: 28-33.
147. Okajima K, Koga S, Okabe H, Inoue M, Takatsuki K. Characterization of the fibrinolytic
state by measuring stable cross-linked fibrin degradation products in disseminated
inn-avascular coagulation associated with acute promyelocytic leukemia. Acta Haematol
1989; 81: !5-18.
148. Fruchtman S. Aledort LM. Disseminated intravascular coagulation. J Am Coll Cardiel
1986: 8: 159-67.
59
149. Sigal SH, _Cembrowski GS, Shatill SJ, Brown NM, Schifreen RS, Schwanz MW.
Prototype quantitative assay for fibrinogen/fibrin degradation products. Clinical
evaluation. Arch Intern Med 1987; 147: 1790-1793.
150. Kush11er I. C-reactive protein and the plasma protein response to tissue injury. In :
Kushner I, Volanakis JE, Gerusz H eds. The phenomenon of the acute phase response.
New York, New York Academy of Science 1982: 39-48.
151. Marder VJ, Marrin EJ, Francis CW, Colman RW. Consumptive thrombohemorrhagic
disorders. In: Colman RE, Hirsh J, Marder VJ, Salzman EW eds. Hemostasis and
Thrombosis. Basic principles and clinical practice. 2-nd ed. Philadelphia: JB Lippincon
1987: 975-1015.
152. Whitaker AN. Rowe EA, Masci PP, Gaffney PJ.ldentification ofD-dimer-E complex in
disseminated intravascular coagulation. Thromb Res 1980; 18: 453-459.
153. Sakai Y. Maeda M, Takei F, Matsumoto T, Nishijima Y, Nakamura K. Determination
ofFDP in human plasma by a novel latex immunoassay. Thromb Res 1988; 50:469-479.
154. Carr JM. McKinney M, McDonagh J. Diagnosis of disseminated intravascular
coagulation. Role of D-dimer. Am J Clin Pathol 1989; 91: 280-287.
!55. Greenberg CS, Devine DV. McCrae KM. Measurement of plasma fibrin D-d.imer levels
with the use of a monoclonal antibody coupled to latex bands. Am J Clin Path 1987; 87:
94-!00.
156. Prisco D, Francalanci I. Giusti E, Scarti L, Bonechi F, Gensini GF. Evaluation of new
methods for the selective measurement of fibrin and fibrinogen degradation products.
Thromb Haemostas 1989; 62: 234 (abstract).
157. Fabbrini N. Hoppensteadt D. Walenga JM. Schumacher H, Fareed J. Diagnostic efficacy
of specific fibrinogen and fibrin related degradation products measured by ELISA
methods. Thromb Haemostas 1989; 62: 236 (abstract).
158. Stibbe J, Gomes M, De Oude A. Management of chronic DIC Evaluation of separate
measurements of fibrin (FbDP) and fibrinogen (FgDP) degradation products concomitant
with fibrin monomers (FM). Thromb Haemostas 1989; 62: 373 (abstract).
159. Tanigawa M, Takagi M, Wada H, Deguchi K, Shirakawa S, Suzuki H, Tanaka I.
Coagulation study before onset of intravascular coagulation. Thromb Haemostas 1989;
62: 370 (abstract).
160. Gaffney PJ, Creighton LJ, Perry MJ, Callus M, Thorpe R, Spitz M. Monoclonal
60
antibodies to crosslinked fibrin degradation products (XL-FDP) I. Characterization and
preliminary evaluation jn plasma. Br J Haematol 1988; 68, 83-90.
161. De Wood MA, Spores J, Notske R, Mouse LT, Burroughss R, Golden MS, Lang HT.
Prevalence of total coronary occlusion during the early hours of transmural myocardial
infarction. N Eng! J Med !980; 303: 897-902.
162. _pavies MJ, Thomas A. Thrombosis and acute coronary artery lesions in sudden cardiac
ischemic disease. N Eng1 J Med 1984; 310: 1137-1140.
163. Gruppo Italiano per lo studio della streptochinasi nell'infarcto miocardia (Gissi). Lancet
1986; 1: 397-401.
164. Koren G, Weiss AT, Hasin Y, Appelbaum D, We1ber S. Rozenman Y, Lotan C, Mosseri
M, Sapoznikov D, Luria MH, Gotsman MS. Prevention of myocardial damage in acute
myocardial ischaemia by early treatment with intravenous streptokinase. N Engl 1 Med
1985: 3!3: 1384-1389.
165. Simoons ML. Van de Brand M, De Zwaans C, Verheugt FWA, Remme WJ, Serruys PW,
Bar F. Res J, Krauss XH, Yenneer F. Improved survival after early thrombolysis in acute
myocardial infarction. Lancet 1985: ii: 578-581.
166. Marder VJ, Sherry S. Thrombolytic therapy: Current status. N Eng! J Med 1988; 318:
1512-!520.
!67. Arnold AER, Brower RW, Collen D, VanEs GA, Lubsen J, Serruys PW, Simoons ML,
Verstraete M. Increased serum levels of fibrinogen degradation products due to treatment
with recombinant tissue-type plasminogen activatOr for acute myocardial infarction are
related to bleeding complications. but not to coronary patency. JAm Coil Cardioll989;
14: 581-588.
168. Francis CW, Connagha_ll DG, Scot WL, Marder YJ: Increased plasma concentration of
cross-linked fibrin poly:ners in acute myocardial infarction. Circulation 1986; 75: 1170-
1177_
!69. Eisenberg PR. Sherman LA, Tiefenbrunn AJ, Ludbrook PA, Sobel BE. Jaffe AS.
Sustained fibrinolysis after administration of t-PA despite its short half-life time in the
circulation. Thromb Haemostas !987; 57: 35-40.
170. Eisenberg PR, Sobel BE, Jaffe AS. Characterization in vivo of the fibrin specificity of
activators of the fibrinolytic system. Circulation 1988; 78: 592-597.
171. Kudryk B, Rohoza A, Ahadi M, Chin J, Wiebe ME. A monoclonal antibody with ability
61
to distinguish between NH2-tenninal fragments derived from fibrinogen and fibrin. Mol
!mmunol 1983; 20; 1191-1200
172. Seifried E. Tanswell P. Ellbruck D. Haerer W, Schmidt A. Pharmacokinetics and
haemostatic status during consecutive infusions of recombinant tissue-type plasminogen
activator in patients with acute myocardial infarction. Thromb Haemostas 1989; 61:497-
50!.
173. Declerck P, Mombaens P, Holvoet P, Collen D. Plasma levels of fragment D-dimer of
cross-linked fibrin during thrombolytic therapy with recombinant tissue -type
plasminogen activator. Thromb Haemostas 1987; 58: 231 (abstract).
174. Brenner CW, Francis CW., Marder VJ. The role of soluble cross-linked fibrin in D-dimer
immunoreactivity of plasmic digests. J Lab Clin Med 1989; 113: 682-688.
175. Eisenberg PR, Sherman LA, Perez J, Jaffe AS. Relationship between elevated plasma
levels of crosslinked fibrin degradation products (XL-FOP) and the clinical presentation
of patients with myocardial infarction. Thromb Res 1987; 46: 109-120.
176. Franzen J. Nilsson B, Johansson BW. Nilsson IM. Fibrinolytic activity in men with acute
myocardial infarction before 60 years of age. Acta Med Scand 1983; 214: 339-344.
177. Estelles A, Tormo G, Azner J. Espana F, Tormo V. Reduced fibrinolytic activity in
coronary heart disease in basal conditions and after exercise. Thromb Res 1985; 40: 373-
383.
178. Hamstein A, Wiman B, De Faire U, Blomback M. Increased plasma levels of a rapid
inhibitor of tissue plasminogen activator in young survivors of myocardial infarction.
N Eng! J Med 1985; 313: 1557-1563.
179. Rogers S, Sweetnam PM, Perry MJ, Gaffney PJ. Plasma levels of fibrin fragments in
men with myocardial infarction. Thromb Res 1986; 43: 389-393.
180. Soria C, Soria J, Me Morshahi, Mirshahi M, Dunnica S, Boucheix C, Beaufils R, Slama
R, Caen JP. Dynamic coronary fibrinolysis evaluation in patients with myocardial
infarction and unstable angina by specific plasma fibrin degradation product
determination. Thromb Res 1987; 45: 383-392.
181. Douglas JT, Lowe GDO, Hillis WS, Rao R, Hogg KJ, Gemmill JD. Endogenous
fibrinolysis in acute myocardial infarction compared to unstable angina. Thromb
Haemostas 1989; 62: 590 (abstract).
182. Brommer EJP, Engbers J, Van de Laarse A, Nieuwenhuizen W. Survival of fibrinogen
62
degradation products in the circulation after thrombolytic therapy for acute myocardial
infarction. Fibrinolysis 1987; 1: !49-153.
183. Kluft C. McNeill A, Adgey AAJ, Rijken DC, Nieuwenhuizen W, Cohen AE. A
randomized dose ranging study of two chain tissue-type plasminogen activator (BW/t-PA)
in myocardial infarction: In vitro monitoring. Thromb Haemostas 1987; 54: (abstract
175).
184. Seifried E, Hoegee-de Nobel E, Haere W, Ellbruck D, Tanswell P, Nieuwenhuizen W.
Intact fibrinogen and tlbrin(ogen) degradation products in the evaluation of fibrin
specificity of tissue-type plasminogen activator. Thromb Haemostas 1989; 62: 300
(abstract).
185. Rocha E, Paramo JA. Fernandez FJ, Cuesta B, Hernandez M, Paloma MJ, Rifon J.
Clotting activation and impairment of fibrinolysis in malignancy. Thromb Res 1989; 54;
699-707.
186. Wojtukiewicz MZ, Zacharski LR, Memoli VA. Kisiel W, Kudryk BJ, Rousseau SM,
Stump DC. Absence of components of coagulation and coagulation and fibrinolysis
pathways in situ in mesothelioma. Thromb Res 1989; 55: 279-284
187. Paramo JA, Cambell W. Cuesta B, Gomez C, Aranda A. Rocha E. Fibrinolytic response
in malignancy. Fibrinolysis 1987: 1: 195-199.
188. Zacharski LR, Memoli VA, Rousseau SM. Cancer-coagulation interaction and in situ in
renal cell carcinoma. Blood 1986: 68: 394-399.
189. Zacharski LR, Memoli VA, Rousseau SM. Thrombin specific sites of fibrinogen in small
cell carcinoma of the lung. Cancer 1988; 62: 299-302.
190. Francis RB Jr, Seyfert U. Tissue plasminogen activator antigen and activity in
disseminated intravascular coagulation: clinico-pathologic correlations. J Lab Clin Med
1987: 110: 541-547.
191. Kirchheimer JC, Pflueger H, Hienen G, Binder GR. Increased urokinase activity to
antigen ratio in human renal-cell carcinoma. Eur J Cancer 1985: 35: 737-741.
192. Grondahl-Hansen J, Agerlin N, Munkholm-Larsen P, Bach F, Nielsen LS,
Dombernowsky P, Dano K. Sensitive and specific enzyme-linked immunosorbent assay
for urokinase-type plasminogen activator and its application to plasma from patients with
breast cancer. J Lab Clin Med 1988; Ill: 42-41.
193. De long E, Knot EAR, Piker D, Iburg AHC, Rijken DC, Veenhof KHN, Dooijewaard
63
G, The Cate JW. Increased plasminogen activator inhibition levels in malignancy.
Thromb Haemostas 1987; 57: 140-143.
!94. Kirchheimer JC, Huber K, Wagner 0, Binder BR. Pattern of fibrinolytic parameters in
patients with gastrointestinal carcinomas. Br J Haematol 1987; 66: 85-89.
!95. Rafter R, Schmitt M, Janicke F, Hollrieder A, Graeff H. Quantitative analysis of
fibrinolytic factors involved in tumor-associated proteolysis of human breast and ovarian
cancer. Thromb Haemostas 1989; 62: 299 (abstract).
!96. Janicke F, Schmitt M, Rafter R, Hollrieder A, Prechtl A, Yon Weiden bach U, Bruaer
H, Graeff H. Tumor-associated fibrinolysis in breast and ovarium cancer- relationship
to clinical data. Thromb Haemostas 1989; 62: 142 (abstract).
197. Baicchi U, gadducci A, Sagripami A, Vispi M, Monellini L, Facchini V, Fioretti P.
Hemostatic activation in gynaecological tumors. Thromb Haemostas 1989; 62: 136
(abstract).
198. Berek JS. Knapp RC. Malkasian GD, Lavin PT. Whitney C, Niloff JM, Bast Jr RC. CA
125 serum levels correlated with second-look operations among ovarian cancer patients.
Obstet Gynecol 1986; 67: 685-689.
199. Nieuwenhuizen W. New strategies in the determination of fibrin and fibrin(ogen)
derivatives by monoclonal antibodies. Blut 1988; 57: 285-291.
200. Wajima T, Mukhopadhyay P. Serial coagulation profiles in patients with small cell
carcinoma of the lung. Thromb Haemostas 1989; 62: 136 (abstract).
64
Chapter 2
CORRELATIONS BETWEEN PLASMA LEVELS OF FIBRIN(OGEN)
DERIVATIVES AS QUANTIFIED BY DIFFERENT ASSAYS BASED ON
MONOCLONAL ANTIBODIES
H. Kroneman'~. W. Nieuwenhuizen3, E.A.R. Knot1
,
P.F.M.M. Van Bergen', M.P.M. De Maat1
Dept. of Internal Medicine II', University Hospital Rouerdarn-Dijkzigt, Center for
Clinical Decision Analysis2, Erasmus University Rotterdam, Gaubius Institute TNO',
Leiden, Thrombosis Service Center' Rotterdam, The Netherlands
Thrombosis Research 1991; 61:441-452
ABSTRACT
New plasma assays for fibrin( ogen) degradation products have become available which
are based upon monoclonal antibodies and can be performed in plasma. In this smdy
we have evaluated four of such specific enzyme immuno assays i.e.: for the total of
degradation products of fibrin and of fibrinogen (TDP), fibrin degradation products (D
dimer and FbDP) and fibrinogen degradation products (FgDP) in patients suspected of
having deep venous thrombosis of the leg (DVT) and patients with cirrhosis of the
liver. In each of the (sub)groups of patients, a very good correlation (0.90< r <0.98)
was observed between the actually measured TDP values and the calculated sum of the
separately measured FbDP and FgDP levels. Only 2% (5 patients) of the cases showed
a discrepancy of more than a factor two between the found TDP values and the
calculated sum of the measured FbDP and FgDP levels.
About 90% of the fibrin degradation products were crosslinked. FbDP levels correlated
well with the FgDP levels (0.72< r <0.94) and D-dimer levels (0.82< r <0.91) in both
patients with DVT and cirrhotics. In those patients also a good correlation (0.67< r
<0.83) was observed between FgDP and D-dimer levels, but not in patients suspected
of having DVT but with a normal lPG test result. Secondary fibrinolysis appeared to
be accompanied by fibrinogenolysis.
66
INTRODUCTION
For at least two decades, the measurement of the products of plasmin digestion of fibrin
and fibrinogen has been •Jsed to assess the extent of fibrino(geno )lysis in several
clinical conditions (I-ll). The levels of FDP's were mostly measured in serum using
a variety of procedures (13). The Thrombo-Wellco test and the tanned red-cell
hemagglutination inhibition assay are well known examples of such serum-based assays
(13). These assays have been evaluated in order to determine their potential both in the
diagnosis and in the monitoring of the effect of treatment of different diseases (1-7).
Since most of these classical assays for FDP assessment are based upon polyclonal
antibodies which cross-react with fibrinogen, their use is limited to serum samples.
Serum~ however, is known to be a notorious source of artifacts and it has been
demonstrated that, in many cases, the currently used assays for FDP assessment in
serum do not reflect the true level of FDP in plasma (14). Falsely positive results can
be expected if residual fibrin or fibrinogen in serum reacts with the antiserum used to
detect FDP (14) or if lysis occurs during serum preparation. Too low or falsely negative
results may be obtained during serum preparation if certain degradation products, e.g.
fragment X, coagulate (15) or become adsorbed to the clot (14,!6). As a consequence,
data based on these serum assays should be interpreted with caution.
Since a few years, new assays have become available for the quantitative determination
of fibrin(ogen) degradation products. These enzyme immunoassays (EIA's) are based
upon specific monoclonal antibodies each of which is directed against specific epitopes
of a certain fibrin derivative. They can be performed on plasma samples. thus avoiding
the serum artifacts. The clinical utility of these assays in several states of diseases has
been reviewed (21).
The currently available EIA' s allow the specific determination of degradation products
of fibrinogen (FgDP) (22), the total of degradation products derived from fibrin and
fibrinogen (TDP) (23) and fibrin degradation products, FbDP (19) and D-dimer (17, 18,
28). Using a combination of these assays it is possible to discriminate between lysis of
fibrinogen and lysis of fibrin (primary and secondary fibrinolysis respectively). It has
been suggested that secondary fibrinolysis is associated with low levels of
67
fibrinogenolysis (30).
Theoretically. it may be expected that the calculated sum of the FbDP and the FgDP
plasma levels will approximate the actually measured TOP values. Such an observation
has been made in patients with ischemic hean disease and patients with disseminated
intravascular coagulation (DlC) (29). Secondary or reactive fibrinolysis is reflected by
thelevels of fibrin degradation products (FbDP), which can be determined by the EIA
for FbDP (which measures both crosslinked and non-crosslinked degradation products)
or (when they are crosslinked by factor XIlla) the D-dimer E!A.
In this study we evaluated these plasma based assays in groups of patients with an
activated state of coagulation. The aims of our study were: !. To examine whether the
measured levels of the total of degradation products of fibrin and fibrinogen (TOP)
correlate with the levels obtained by calculation of the sum of the separately measured
FbDP and FgDP levels. 2. To compare the levels of fibrin degradation products,
assessed by the D-dimer E!A. with the levels of fibrin degradation products, as
determined by the EIA for FbDP. 3. To see whether secondary fibrinolysis is associated
with fibrinogenolysis by investigating the relationship between FgDP levels and the
levels of fibrin degradation products, as detected by the EIA's for FbDP and D-dimer.
PATIENTS AND METHODS
The following groups of subjects were studied:
- group 1 consisted of 232 consecutive outpatients clinically suspected of having deep
venous thrombosis (DVT). which have been described before (24). They were referred
to the Thrombosis Service Center by their general practitioner. The male: female ratio
was 63:169. the mean age 67 (range:l7-94) and the mean duration of symptoms 9.5
days. DVT was assessed by impedance plethysmography (lPG) and the results were
interpreted according to Hull's criteria (25). The prevalence of DVT in this outpatients
population was 25% (59 out of 232 patients).
- group 2 consisted of 31 hospitalized patients with cirrhosis of the liver as confirmed
by liver biopsy. All patiems were admitted to the hospital for analysis of the severity
of their liver disease. Based on the Child-Pugh classification (26), the patients were
68
divided in 13 patients with Child A, 9 patients with Child B and 9 patients with Child
C. The mean age was 49 (range:3!-70) and the male:female ratio was 13:18.
- group 3: The reference group consisted of 43 apparently healthy hospital employees
(age range: 22-45).
In all cases, informed consent was obtained and the research was earned out according
to the principles of the Declaration of Helsinki.
Blood samples
Venous blood samples were collected by venepuncture into vacuum tubes containing
cold trisodiumcitrate 0. II mol/] (I volume per 9 volumes blood). Plasma was obtained
by centrifugation for 30 minutes at 2000 x g at 4' C. The plasma samples were stored
frozen in small aliquots at -70'C until tested and carefully thawed at 37'C prior to the
start of the assay.
Assays
All assays were performed on plasma samples in duplicate. The three used assays for
the specific determination of the total of degradation products of fibrin and fibrinogen
(TDP: Fibrinostika TDP). fibrin degradation products (FbDP; Fibrinostika FbDP) and
fibrinogen degradation products (FgDP; Fibrinostika FgDP) were enzyme
immunoassays of the sandwich type (Organon Teknika, Tumhout, Belgium). All three
test kits are based upon the same specific monoclonal antibody as capture antibody
(FDP-14). FDP-14 has its epitope in theE-domain of the fibrinogen molecule i.e. on
the BB-chain between amino-acids 54-118. It reacts both with degradation products of
fibrin and of fibrinogen, but not with the intact parent molecules (27). For the specific
separate determination of fibrinogen degradation products (FgDP's) in the EIA for
FgDP, a horse-radish peroxidase (HRP) conjugated monoclonal antibody (Y-18) is used
as tagging antibody which reacts with fibrinopeptide A-comprising fibrinogen fragments
(22). The tagging antibody for the specific FbDP assay is a HRP conjugated
monoclonal antibody (DD-13) against fibrin degradation products (crosslinked and non
crosslinked) (19). In the TDP assay kit for determination of degradation products of
69
fibrin and fibrinogen, a mixture of the two monoclonals Y-18 (22) and DD-13 (19) is
used as tagging antibody (27). D-dimer was determined by an enzyme-linked
immunosorbent assay (Boehringer Mannheim, West-Germany) using a monoclonal anti
FDP-D.D. antibody as capture antibody, which strongly reacts with a neoantigenic
determinant in D-dimer and has some cross reactivity with fragment D (28). The
tagging antibody is a panspecific anti-FDP-D peroxidase-conjugated antibody.
Statistical analysis
The difference between the means of the test results in the different groups of patients
was determined using the Wilcoxon ranksum test (two tailed). Pearson's correlation
coefficient was calculated to determine the relationship between the plasma levels of
fibrin(ogen) degradation products by using linear regression analysis.
RESULTS AND DISCUSSION
The median and range of the plasma levels (in ug/ml) of TDP, FbDP, FgDP and D
dimer in the different groups of subjects are shown in table I.
Table I.
group I group 2 group 3 (n=232) (n=31) (n=43)
FbDP 0.55 0.61 0.19 (0.17-14.6) (0.21-8.2) (0.17-0.45)
FgDP 0.36 0.34 0.19 (0.17-9.6) (0.23-1.27) (0.17-0.29)
IDP 0.88 0.97 0.42 (0.21-23.1) (0.29-!0.3) (0.2!-0.65)
D-dimer 0.66 1.1 0.14 (0.08-13.7) (0.12-8.0) (0.05-0.33)
The median a.~d range of the plasma concentrations (in ug/ml) of FbDP, FgDP,
TDP and D-dimer in patients suspected for DVT (group !),patients with
cirrhosis of the liver (group 2) and norrnal healthy controls (group 3).
70
In both groups of patients, those suspected of having DVT and in the group of patients
with cirrhosis of the liver, the plasma levels of FbDP, FgDP, TDP and D-dimer were
higher than the corresponding plasma levels in normal healthy controls (in all cases
P<O.OOO!). Among the group of patients suspected of having DVT, those with
confirmed DVT had elevated plasma levels (p<O.OOO!) of fibrin(ogen) degradation
products when compared to the levels in patients with a normal lPG test result.
In this study we have evaluated the correlation between the plasma levels of
fibrin(ogen) derivatives, as determined by EIA's for FbDP, D-dimer, FgDP and TDP
in patients suspected of having DVT and cirrhotic patients. The correlation coefficients
of the fibrin(ogen) degradation products are shown in table 2.
suspected DVT con!rrmed DVT noDVT cirrhosis normals
(n=232) (n=59) (n=l73) (n=31) (n=43)
TDP versus FgDP + FbDP 0.94 0.92 0.90 0.98 0.401)
FbDP versus D-dimer 0.81 0.82 0.35 0.91 0.35"
FgDP versus FbDP 0.93 0.94 0.45 0.72 0.65
FgDP versus D-dimer 0.81 0.83 O.QI~ 0.67 0.39"
All P values were< 0.0001, except: 1) p < 0.05; 2) p ::::: ns
Table 2. Correlation coefficients of fibrin(ogen) degradation products levels in different
(sub)groups of patients.
In the patients suspected of having DYT (fig 1) and patients with cirrhosis of the liver
(fig 2). the actually measured plasma levels of the total of fibrin and fibrinogen
degradation products (EIA for TDP) showed a very good correlation (0.90< r <0.98,
p<O.OOO!) with the values calculated by adding the measured levels of FbDP to the
71
levels of FgDP. This obs~rvation is in accordance with observations by others in
patients with ischemic heart disease and DIC (29). The correlation between TDP and
the sum of FbDP and FgDP did not differ much in the subgroup of patients with
confirmed DVT (r=0.92) and patients with a normal lPG test result (r=0.90). The
correlation between the measured TDP values and the calculated sum of the measured
FbDP and FgDP levels in the whole group of patients suspected of having DVT is
shown in figure I.
TOP (ug/ml) 100~--~--------------------------------------~
10
.. • -.·· iC..
1 ·. :!~ ... .
.
'· Y • 1.17X - 0.237 r • 0.94
n • 232
0.1L---~~~~LL~----~-L-L~LL~----~~_L~_u~
0.1 1 10 100 calculated sum of FbDP • FgDP (ug/ml)
Fig 1. Correlation between t!1e measured TDP values and the calculated sum of the FbDP and
the FgDP levels (in ug!ml) in patients clinically suspected of having DVT (n=232).
In five of the 232 patients suspected of having DVT, we found a discrepancy of more
than a factor two between the measured TDP levels and the calculated sum of the
FbDP and FgDP levels . The ratio (FbDP+FgDP)(fDP was higher than 2 in four
subjects who happened to have an lPG examination with a normal test result. In one
72
patient with an abnormal lPG test result, the ratio was only 0.13. In this latter case, it
means that the plasma levels of fibrin(ogen) degradation products obtained by the EIA
for TDP was 7.88 times higher than the values obtained by calculating the sum of the
FbDP and the FgDP levels. The reason for this is not clear. None of the patients with
cirrhosis of the liver nor the healthy controls had (FgDP+FbDP)ffDP ratios above 2.0
or below 0.5.
T ·=D~P~(=ug~/~m~l)~-------------------------------------, 100 <=
y • 1.18X - 0.08
r • 0.98
n • 31
..
0.1 L_ __ .L___.J_.,L_.,L_j_j___L_l_j_ __ __j__.,L__L_L__j_.LJ__LJ
0.1 1 10 Calculated sum of FbDP + FgDP (ug/ml)
Fig 2. Correlation between the measured TDP values and the calculated sum of the FbDP and
the FgDP levels (in ug/ml) in patients with cirrhosis of the liver (n=31).
Secondly, we investigated the correlation between the plasma levels of fibrin
degradation products as measured by the D-dimer EIA and the ElA for FbDP. A good
correlation was observed between D-dimer values and FbDP values (both of which
reflect secondary fibrinolysis) in patients with confirmed DVT and patients with
cirrhosis of the liver (0.82< r <0.92, p<O.OOOI).
The correlation coefficient for the FbDP and D-dimer values was r = 0.81 (p<O.OOOl)
73
in all patients suspected of having DVT (see figure 3). This correlation was the about
same in the subgroup of pctients with confirmed DVT (r=0.82; p<O.OOOl) and poor in
the subpopulation of patients with a negative lPG test result (r=0.35; p<O.OOOI).
FbDP (ug/ml) 100~--~---------------------------------------,
1
y • o.ssx. 0.11
r • 0.81
n 232
, . .. . . . . . . .
.. · ..... ·;!:-:.· . . . ~. ·"· .. -~-· ~ . , . . , . .... . .... . -.·'!.. :.-::::- ... -. ..~ ., .. _
• , .r • • :_:.· -:: ...
.·
0.1~~~~~~~~LWW-~-LLU~--~LLLUilL~-L~~
0.001 0.01 0.1 1 D-dimer (ug/ml)
10 100
Fig 3. Correlation between plasma levels of FbDP and D-dimer (in ug/ml) in patients
suspected of having DVT.
The correlation between FbDP and D-dimer in patients with cirrhosis of the liver is
shown in figure 4.
In contrast to the EIA for D-dimer. which only detects crosslinked fibrin degradation
products, the FbDP EIA measures both crosslinked and non-crosslinked fibrin
derivatives. When one assumes that the calibration materials in the FbDP and the D
dimer EIA are comparable. one can conclude from the equations of the regression lines
in figure 3 and 4, that about 90% of the measured FbDP levels consisted of D-dimer.
This suggests that nearly all of the measured fibrin degradation products are crosslinked
in those patients. We are aware of only one study that has described a correlation
74
(r=0.72, p<O.Ol) between D-dimer values and FbDP levels (30). However, this
concerned a different
D-dimer EIA using another monoclonal antibody (DD-386/22).
The third purpose of our study was to examine whether secondary fibrinolysis is
accompanied by fibrinogenolysis. In each of the (sub)groups of patients, the FgDP
levels correlated less with D-dimer values than with FbDP levels (table 2). The
observation of positive correlations between plasma levels of fibrin degradation
products, as assessed by EtA's for FbDP and D-dimer, and the FgDP values confmns
earlier observations that secondary fibrinolysis seems to be accompanied by lysis of
fibrinogen (30). Interestingly, both the correlations between FgDP and FbDP and
between FgDP and D-dimer levels were higher in the reference group (r=0.65, r=0.39
respectively) than in the group of patients with a negative lPG test (r=0.45, r=O.Ol
respectively). We have no explanation for the lack of correlation (r=O.Ol, p=ns)
between D-dimer and FgDP levels in patients with a negative result of IPG
examination. This observation, however, does not necessarily imply that there is no
relationship, but rather that there is no linear relationship.
In general. the correlations between the fibrin(ogen) degradation products were lower
in normals than in the different (sub)groups of patients. This may be explained by the
fact that the fibrin(ogen) concentrations in normals are generally very low. Following
the manufacturers instructions for dilutions of the samples, this means that the
concentrations have to be read from the lower (nearly horizontal) part of the sigmoidal
shaped calibration curves, and this yields inaccurate results. When this study was
finished, we learned that lower dilution factors than 20-fold (as prescribed by the
manufacturer) are allowed. This might improve the correlations. Summarizing, we
found that: 1. The levels of the total degradation products (TDP EIA) show a very good
correlation with the calculated sum of the measured FbDP and the FgDP levels. 2.
There is a good correlatio~ between plasma D-dimer and FbDP levels and most of the
fibrin degradation products were crosslinked. 3. Secondary fibrinolysis appeared to be
associated with fibrinogenolysis.
75
FbDP (ug/ml) 10c---~~-----------------------------------,
Y • 0.91X - 0.26
r • 0.91
n • 31
0.1L_ ____ L_~L_~J_~~~-----L--~~~-L~~
0.1 1 D-dimer (ug/ml)
10
Fig 4. Correlation between FbDP and D-dimer plasma levels (ug/rnl) in patients with cirrhosis
of the liver (n=31).
REFERENCES
I. Ruckley C V. Das PC. Leitch A G. Donaldson A A. Copland W A, Redpath
AT. Scott P. Cash J D. Serum fibrin/fibrinogen degradation products associated
with postoperative pulmonary embolus and venous thrombosis. Br Med J, 4,
395-398, 1970
2. Hedner U, Nilsson I M. Clinical experience with determination of fibrinogen
degradation products. Acta Med Scand, 189,471-477, 1971
3. Wood E H, Prentice C R M, McNicol G P. Association of fibrinogen- fibrin
related antigen (F.R. antigen) with postoperative deep vein thrombosis and
systemic complications. Lancet, i, 166- 169, 1972
4. Gallus A S, Hirsch J, Gent M. Relevance of preoperative and postoperative
blood tests to postoperative leg vein Thrombosis. Lancet, ii, 805 - 809, 1973
76
5. Gurewich V. Hume M, Patrick M. The laboratory diagnosis of venous
thromboembolic disease by measurement of fibrinogen/fibrin degradation
products and fibrin monomer. Chest, 64, 585-590, 1973
6. Cooke E D, Gordon Y B, Bowcock S A, Sola S A, Pilcher F M, Chard T,
Ibbotson R M, Ainsworth M E. Serum fibrin(ogen) degradation products in
diagnosis of deep vein thrombosis and pulmonary embolism after hip surgery.
Lancet, ii, 51-54, 1975
7. Clayton J K, Anderson J A, McNicol G P. Preoperative prediction of
postoperative deep vein thrombosis. Br Med J, 2, 910-912, 1976
8. Leslie J, Langler D, Serjeant G, Serjeant B, Desai P, Gordon Y. Coagulation
changes during the steady state in homozygous sickle-cell disease in Jamaica.
Br J Haematol, 30, 159-166, 1975
9. Gaffney P J, Perry M J. "Giant" fibrin fragments and thrombosis. Thromb
Haemostas, 54, 93 I, I 985
I 0 Gaffney P J, Creighton L C, Harris R, Perry M 1. Monoclonal antibodies
(MABS) to crosslinked fibrin fragments. Their characterization and potential
clinical use. in: Miiller-Berghaus G, Scheefers-Borchel U, SelmayrE, Henschen
A, eds. Fibrinogen and its derivatives. Biochemistry, physiology and
pathophysiology. Amsterdam. Excerpta Medica, !986. pp. 273-284.
II Bick R L. Disseminated intravascular coagulation: a clinical/laboratory study of
48 patients. Ann NY Acad Sci, 370, 843, l98I
12 Donati M B. Assays for fibrinogen/fibrin degradation products in biological
fluids: some methodological aspects. Thromb Diath Haemorrh, 34, 652 - 662,
1975
13 Merskey C, Lalezazi P, Johnson A 1. A rapid simple sensitive method for
measuring fibrinolytic split products in human serum. Proc Soc Exp Bioi Med
NY, 131, 87I-875, I969
14 Gaffney P J, Perry M J. Unreliability of current serum fibrin degradation
products (FDP) assays. Thromb Haemostas, 53, 301-302, 1985
15 Marder V J, Shulman N R. High molecular weight derivatives of human
77
fibrinogen produced by plasmin II. Mechanism of their anticoagulant activity.
J Bioi Chern, 244,2120-2124, 1969
16 Niewiaroski S, Stewart G 1, Marder V 1. Formation of highly ordered polymers
fibrinogen and fibrin degradation products. Biochim Biophys Acta, 221, 326-
341, !970
17 Elms M J, Bunce 1 H, Bundesen P G, Rylan D B, Webber A J, Masci P P,
Whitaker A N. Measurement of crosslinked fibrin degradation products - An
immunoassay using monoclonal antibodies. Thromb Haemostas, 50, 591-594,
1983
18 RylanD B, Blake A S, Glottis L E, Massingham D A, Fletcher W A, Masci P
P. Whitaker AM, Elms M, Bunce J, Webber A, Wyatt D, Bundesen P G. An
immunoassay for human D-dimer using monoclonal antibodies. Thromb Res, 31,
767-778, 1983
19 Koppen P W. Hoegee-de Nobel E, Nieuwenhuizen W A. A monoclonal
antibody-based enzyme immunoassay for fibrin degradation products in plasma.
Thromb Haemostas. 59, 310-315, 1988
20 Whitaker AN, Elms M J. Masci P P, Bundesen P G, RylanD B, Webber A J,
Bunce I H. Measurement of cross linked fibrin derivatives in plasma: an
immunoassay using monoclonal antibodies. J Clin Path. 37, 882-887, 1984
21 Kroneman H. Nieuwenhuizen W, Knot E A R. Monoclonal antibody-based
plasma assays for fibrin(ogen) and derivatives, and their clinical relevance.
Blood Coagulation and Fibrionolysis, I, 91-111, 1990
22 Koppen P W. Kuipers W, Hoegee-de Nobel E, Brommer E 1 P, Koopman J,
Nieuwenhuizen W. A quantitative enzyme immunoassay for primary
fibrinogenolysis products in plasma. Thromb Haemostas, 57, 25-28, 1987
23 Koopman J, Haverkate F, Koppen P W, Nieuwenhuizen W, Brommer E J P,
Van der Werf W G C. New enzyme immunoassay of fibrin-fibrinogen
degradation products in plasma using a monoclonal antibody. 1 Lab Clin Med,
109, 75-84, !987
24 Van Bergen P F M M, Knot E A R, Jonker J J C, De Boer A C, De Maat M
78
P M. Is quantitative determination of fibrin( ogen) degradation products and
thrombin-antithrombin Ill complexes useful to diagnose deep venous thrombosis
in outpatients? Thromb HaemostaS, 62, 1043-1045, 1989
25 Hull R, van Aken W G, Hirsch J, Gallus AS, Hoicka G, Turpie A G G, Walker
I, Gent M. Impedance Plethysmography using the occlusive cufftechnic in the
diagnosis of DVT. Circulation, 53, 696-700, 1976
26 Pugh R N H, Murray-Lyon I M, Dawson J L, Pietroni M C, Williams R.
Transection of the eosophagus for bleeding oesophageal varices. Br J Surg, 60,
646-649, 1973
27 Koppen P W, Koopman 1, Haverkate F. Nieuwenhuizen W. Production and
characterization of a monoclonal antibody reactive with a specific neoantigenic
determinant (comprising BB 54-118) in degradation products of fibrin and of
fibrinogen. Blood, 68,437-441, 1986
28 Amiral J, Plassart V, Minard F. Measurement and clinical relevance of D-dimer
by ELISA. ln:Fibrinogen and its derivatives. Muller-Berghaus G, Scheefers
Borchel U. Selmayr E, Henschen A (Eds.) Amsterdam-New York-Oxford:
Excerpta Medica 1986, pp. 285-290
29 Prisco D, Pannicia R, Bonechi F, Francalanci I, Abbate R, Gensini G F.
Evaluation of new methods for the selective measurement of fibrin and
fibrinogen degradation products. Thromb Res, 56, 547-551, 1989
30 Nieuwenhuizen W. A double blind comparative study of six monoclonal
antibody-based plasma assays for fibrinogen derivatives. In: Lowe GDO,
Douglas JT, Forbes CD, Henschen A. eds. Fibrinogen 2. biochemistry,
physiology and clinical relevance. Amsterdam, Excerpta Medica 1987, pp. 181-
186.
79
Chapter 3
DIAGNOSTIC VALUE OF D-DIMER FOR DEEP VENOUS THROMBOSIS
IN OUTPATIENTS
Herman Kroneman'~, Paul F.M.M. Van Bergen', Eduard A.R. Knot',
Jan J.C. Jonker', Moniek P.M. de Maar'
Center for Clinical Decision Analysis', Erasmus University Rotterdam, Departtnent of
Internal Medicine 112, University Hospital Rotterdam-Dijkzigt, Thrombosis Service
Center'. Rotterdam. The Netherlands
Haemostasis !99!;21:286-292
ABSTRACT
We have studied the diagnostic value for deep venous thrombosis (DVT) of an enzyme
immunoassay (ElA) for the detection of D-dimer in plasma of 239 consecutive out
patients suspected of having DVT by their general practitioner. DVT was conf"mned by
impedance plethysmography in 60 patients. Using the 95"' percentile range of 42
healthy volunteers, the sensitivity for the detection of DVT was 92%, with a specificity
of 21%. In our population with a prevalence of 25%, the D-dimer E!A showed a
negative predictive value of 88% and a positive predictive value of 28%. We conclude
that this D-dimer ELISA has limited value, either to confmn or to exclude DVT in out
patients.
82
INTRODUCTION
The diagnosis of deep venous thrombosis (DVT) based upon clinical signs and
symptoms only, is confirmed by objective testing in less than 50% of the cases [l-3].
In order to avoid unnecessary exposure to the risk of haemorrhagic complications due
to anticoagulant treatment [4], the need for confirmative diagnosis by objective methods
is widely accepted. The gold standard for the diagnosis of DVT, contrast venography
[5], is an invasive, often painful and expensive technique that requires special expertise
[6]. Therefore several non-invasive alternatives have been developed, of which
impedance plethysmography (lPG) [7-10], ultrasonography [II, 12] and 1251-fibrinogen
leg scanning [13] have extensively been evaluated. lPG has shown to be a practical
non-invasive alternative for the diagnosis of proximal DVT of the leg in outpatients [8-
!0].
In the past, laboratory assays have also been investigated to establish their potential role
in the diagnosis of DVT [14-20]. Most of these assays were based upon the idea that
the presence of products in the circulation, resulting from plasmin mediated lysis of
fibrin(ogen), might be a useful marker of a thrombotic event. Because these assays
employed polyclonal antibodies which crossreact with fibrinogen, their use is limited
to serum samples. It has been demonstrated, however, that results obtained by these
serum-based assays are unreliable [21]. Other laboratory assays focused upon the
assessment of substances generated during thrombus formation, such as fibrinopeptide
A and B-thromboglobuline. but with disappointing results [22-24].
Assays have now become available for the detection of fibrin(ogen) degradation
products in plasma [25-30]. They can be divided in (semi-quantitative) latex
agglutination assays and (quantitative) enzyme immunoassays (E!A). These assays are
based upon monoclonal antibodies with a well defined specificity directed against
epitopes of a certain fibrin(ogen) derivative and can be performed in plasma samples.
The D-dimer EIA is a well known representative of these plasma assays. Plasmin
induced lysis of cross-linked fibrin produces derivatives containing D-dimer fragments,
i.e. two covalent! y bound D-domains [31] and fragment E. Therefore the presence in
blood of circulating fibrin derivatives containing D-dimer are specific markers of a
83
thrombotic process and its >ubsequent reactive fibrinolysis in vivo. The purpose of this
study was to determine the rest characteristics of the D-dimer EIA for the diagnosis of
deep venous thrombosis in outpatients.
PATIENTS AND METHODS
The study comprised consecutive outpatients clinically suspected of having DVT who
were routinely referred to the Thrombosis Service Center Rotterdam by their general
practitioner. The same group of patients has been used to determine the test
characteristics of the assays for Thrombin-antithrombin Ill (TAT) complexes.
crosslinked and non-crosslinked fibrin degradation products (EIA for FbDP). fibrinogen
degradation products (E!A for FgDP) and the total degradation products of fibrin and
fibrinogen (EIA for TDP) [32]. Informed consent was obtained from each patient before
lPG was performed and a venous blood sample was drawn. Blood samples were coded
and handled independently from the lPG result. All lPG results were interpreted by one
of the authors according to the criteria described by Hull [33].
The study was carried out according to the principles of the Declaration of Helsinki.
Venous blood samples were obtained from the antecubital vein into vacuum tubes con
taining nisodiumcitrate 0.11 mol/! (9:1). Within 2 hours after collection. blood was
centrifuged at 1500 x g for 10 minutes at IO'C. Plasma was then collected and stored
frozen in small aliquots at -70°C. and carefully thawed prior to the assay. Venous blood
samples of forty-two oste.Jsibly healthy hospital employees (age 22-35 years) were
handled identically and tested for normal reference values.
D-dimer was determined in plasma (in duplicate) by a sandwich type EJA (Boehringer
Mannheim. Mannheim. Germany). using a specific monoclonal antibody (anti-FDP
D .D.) as capture antibody. which strongly reacts with a neoantigenic determinant in D
dimer and to a lesser extent with fragment D [34]. The tagging antibody was a
panspecific anti-FDP-D peroxidase conjugated antibody.
Statistical analvsis
Differences in results of ptient groups have been tested by means of the Wilcoxon
84
Rank Sum test (two tailed). Any probability less than 0.05 was considered to represent
a significant difference between the studied samples.
RESULTS
A total of239 consecutive outpatients were entered into the study. DVTwas confirmed
by lPG in 60 patients; in 10 of these patients the lPG convened to abnormal in the
second examination. one week after the first examination. Seven patients were excluded
because of improper handling of the plasma samples. The basic characteristics of the
total of 232 patients that remained for analysis are shown in table I. Of these, 59
patients (25%) had an abnormal lPG examination of whom 27% (!6 subjects) had a
prior history of DYT. The median delay between onset of symptoms and the first lPG
examination was 8 days in lPG positive patients and 10 days in patients with a normal
lPG test result. This difference was statistically not significant (p=0.24). No relationship
was found between D-dimer levels and duration of symptoms in lPG negative patients
(p=0.03, p=0.67) or lPG postive patients (p=O.l6. p=0.23).
Within patients that were clinically suspected for DVT, the median D-dimer plasma
levels in patients with a normal lPG examination was 528 (range; 10-5172) ng/ml, while
those with a abnormal lPG examination had a median D-dimer level of 1236 (range:6-
16000) ng/ml. The median D-dimer plasma levels in 42 healthy volunteers (the
reference group) were 148 (range: 53-330) ng/ml (Table II). The cut off point for test
positivity (268 ng/ml) of the D-dimer EIA was obtained by using the 95~ percentile of
the measured D-dimer plasma levels of the normal reference group.
Mean plasma D-dimer values were higher (p<O.OOOJ) in patients with DVT. as
confirmed by an abnormal lPG. when compared with the values in symptomatic non
DVT patients. as determined by a normal lPG examination. Funhermore. symptomatic
non-DYT patients had elevated (p<O.OOOJ) D-dimer levels in comparison to the
reference group.
As can be seen in table Ill. 92% of the patients with DVT had D-dimer levels above
the cut off point for test positivity, leaving 5 cases with a false negative result. Among
the 173 patients. in who DVT could not be confirmed. there were 137 cases
85
Table I. Characteristics of consecutive outpatients with suspected deep venous thrombosis.
N Male/Female Age* Prior history of DVT Duration of syrnptoms*(in days)
* Median and range
IPG abnormal
59 23136 74 (23-94) 6 (27%) 8 (1-122)
IPG normal
173 401133 64 (17-89) 34 (20%) 10 (1-700)
Table II. Test results. of plasma samples of consecutive outpatients with suspected DVT and of the reference group.
Tests Abnormal IPG Normal IPG Reference group
1236 (6-16000) 528 (10-5172) 148 (53-330)
* median and range in ng/ml
Table Ill. Dichotome table showing the results of D-dimer Elisa versus results of IPG examination in consecutive outpatients with suspected DVT
lPG
Abnormal Normal
D-dirner~268 nglml 54 137 191 *PV +:54/ 191=28%
D~dimer<268 nglml 5 36 41 *PV -: 36/ 41=88%
59 173 232
Sensitivity Specificity
54/59=92% 361173=21%
* PV +: predictive value of a positive test
PV-: predictive value of a negative test
86
with a false positive result. Therefore the sensitivity and specificity of the D-dimer EIA
were 92% and 21% respectively. The prevalence of DVT was 25% and the predictive
value of a positive and negative test result were 28% and 88% respectively.
DISCUSSION
The introduction of assays based on monoclonal antibodies, for the detection of
intermediate or endproducts of fibrinolysis in plasma, has raised new interest to
investigate the value of laboratory assays in the diagnosis of several diseases involving
the haemostatic system.
In this study we have measured the test performance of the D-dimer EIA in outpatients
who were clinically suspected of having DVT. The D-dimer assay had a sensitivity of
92%: five of all patients with DVT. confirmed by lPG, had D-dimer levels below the
cut-off point for test positivity (false negative rate of 8%). The specificity was low
(21 %). The test characteristics (sensitivity and specificity) are considered to be
independent of the prevalence of disease (or pretest probability). However, for
application of the D-dimer assay in daily clinical practice, it is relevant to know
whether a test result can predict the presence or absence of disease. Therefore it is
imponant to be aware of the positive or negative predictive value (or post test
probability) of a given test result.
In our population, the prevalence of DVT was 25%. A positive D-dimer test was
followed by an abnormal lPG test in only 28% of the cases (i.e. the positive predictive
value is 28% ). As a consequence, the observation of elevated D-dimer levels alone in
patients suspected of having DVT is not sufficient to indicate anticoagulant treatment.
The predictive value of a negative test result was 88%. Based on a D-dimer test result
only, the diagnosis DVT will be missed in 12% of the cases with D-dimer levels below
the cut-off point for test positivity. This is an unacceptably high percentage.
The sensitivity of the D-dimer EIA in our study is comparable with results reponed by
others using ascending venography [35-38], but the specificity is lower. This Iauer may
be explained by the fact that we used lPG to confirm DVT, which has a limited ability
to detect calf vein thrombosis. The use of ascending venography may lead to a shift of
87
lPG negative patients with elevated D-dimer levels due to calf vein thrombosis to a
DVT positive classification. which results in a higher specificity. It has been
demonstrated. however. that safe management with the use of lPG is possible in
symptomatic outpatients [9].
Application of a different cut-off point for test positivity may also influence the test
results. ln our study it was not possible to reach a l 00% negative predictive value by
variation of the cut-off point. Furthermore, a difference of prevalence of DVT may alter
the predictive value of a test result: e.g. when the prevalence decreases, the positive
predictive value will also fall, but the negative predictive value may rise.
Other recently developed laboratory tests. like the assays for thrombin-antithrombin
(TAT) complexes [32.39] and the assays for cross-linked and non cross-linked fibrin
degradation products (FbDP EIA), for the total of fibrin and fibrinogen degradation
products (TOP EIA) and for fibrinogen degradation products (FgDP ElA) have also
been evaluated in the diagr.osis of DVT [32]. Neither of these assays reached a negative
predictive value of 100%, although it should be noticed that the test characteristics of
the EIA 's for FbDP. TDP and FgDP has not yet been determined using ascending
venography in patients suspected of having DVT.
Other studies, using a semiquantitative D-dimer latex agglutination assay, demonstrated
that DVT could not be excluded if only based on D-dimer levels below the cut-off
point for test positivity [35-37]. Formerly developed assays for determination of
fibrinopeptide A (FPA) and betathromboglobulin (beta-TG) in plasma have also been
evaluated in the diagnosis of DVT [22-24] and have shown to be of little value in the
diagnosis of DVT.
Summarizing, the results cf this study seem to indicate that the D-dimer E!A can not
be used in our population vf outpatients clinically suspected of having DVT to select
patients for further diagnostic procedures (lPG).
Acknowledgements
We are indebted to the Camelis Visser foundation for their financial support,
Boehringer Mannheim. Mannheim Germany for supplying us with the D-dimer ELISA
88
and all the employees of the Thrombosis Service Center Rotterdam for their kind
assistance.
REFERENCES
!. O'Donnell TF Jr, Abbott WM, Athanasoulis CA, Mlllan VG. Caloow AD:
Diagnosis of deep venous thrombosis in the outpatient by venography. Surg
Gynecol Obstet !980; 150:69-74.
2. Cranley JJ, Canas AJ, Sui! WJ: The diagnosis of deep venous thrombosis:
fallibility of clinical symptoms and signs. Arch Surg 1976; Ill :34-36.
3. Haeger K: Problem; of acute deep venous thrombosis. L The interpretation of
signs and symptoms. Angiology 1969; 20:219-232.
4. Levine MN. Raskob G. Hirsch J: Haemorrhagic complications of long-term
anticoagulant therapy. Chest 1986; 2(suppl):16s-25s.
5. Rabinov K. Paulin S: Roentgen diagnosis of venous thrombosis in the leg. Arch
Surg 1972: 104:134-144.
6. Beumann MA. Paulin S: Leg phlebography: the incidence. nature and
modification of undesirable side effects. Radiology 1977; 122:101-104.
7. Huisman MY. Buller HR. Ten Cate JW, Heijerrnans HSF. VanderLaan J, Van
Maanen DJ: Management of clinically suspected acute venous thrombosis with
serial impedance plethysmography in an urban hospital. Arch Intern Med 1989;
149:511-513.
8. Hull RD. Hirsch J. Carter CJ. Jay MR. Ockelford PA, Buller HR, Turpie AG,
Powers P, Kinch D. Dodd PE, Gill GJ, Leclerc JR, Gent M: Diagnostic efficacy
of impedance plethysmography for clinically suspected deep-vein thrombosis:
a randomized trial. Ann Intern Med 1985; 102:21-28.
9. Huisman MY, Buller HR. Ten Cate JW. Vreeken J: Serial impedance
plethysmography for suspected deep venous thrombosis in outpatients: the
Amsterdam general practitioners study. N Eng! J Med 1986: 314:823-828.
10. Huisman MY. Bulkr HR. Ten Cate JW: Utility of impedance plethysmography
in the diagnosis ot recurrent deep-vein thrombosis. Arch Intern Med 1988;
89
148:681-683.
11. Appelman PT, DeJong TE, Lampmann LE: Deep venous thrombosis of the leg:
US findings. Radiology 1987: 163:743-746.
12. Lensing AWA, Prandoni P, Brandjes D, Huisman PM, Vigo M, Tomasella G,
Krekt J, Ten Cate JW, Huisman MY, Buller HR: Detection of deep-vein
thrombosis by real time B-mode ultrasonography. N Eng! J Med 1989; 320:342-
345.
13. Kakkar VV, Nicolaides AN, Renney JTG, Friend JR, Clarke MB: 1251-labelled
fibrinogen test adapted for routine screening for deep-vein thrombosis. Lancet
1970: I :540-542.
14. Ruckley CV. Das PC. Leitch AG,Donaldson. AA, Copland WA, Redpath AT,
Scott P. Cash JD: Serum fibrin/fibrinogen degradation products associated with
postoperative pulmonary embolus and venous thrombosis. Brit Med J 1970;
4:395-398.
15. Redner U, Nilsson I M: Clinical experience with determination of fibrinogen
degradation products. Acta Med Scand 1971; 189:471-477.
16. Wood EH. Prentice CRM, McNicol GP: Association of fibrinogen - fibrin
related antigen (F.R. antigen) with postoperative deep vein thrombosis and
systemic complications. Lancet 1971; i:l66-169.
17. Gallus AS. Hirsch J, Gent M: Relevance of preoperative and postoperative blood
tests to postoperative leg vein Thrombosis. Lancet 1973: ii:805-809.
18. Gurewich V. Hume M. Patrick M: The laboratory diagnosis of venous
thromboembolic disease by measurement of fibrinogen/fibrin degradation
products and fibrin monomer. Chest 1973; 64;585-590.
19. Cooke ED, Gordon YB. Bowcock SA. Sola CM, Pilcher MF. Chard T, Ibbotson
RM, Ainsworth ME: Serum fibrin(ogen) degradation products in diagnosis of
deep vein thrombosis and pulmonary embolism after hip surgery. Lancet 1975;
ii:51-54.
20. Clayton JK. Anderson JA. McNicol GP: Preoperative prediction of postoperative
deep vein thrombosis. Brit Med 1 1976; 2:910-912.
90
21. Gaffney PJ, Perry MJ: Unreliability of current serum fibrin degradation
products(FDP) assays. Thromb Haemost 1985; 53:301-302.
22. Wojciechowski J, Olausson M, Karsan-Bengtsen K: Fibrinopeptide A, B
Thromboglobulin and Fibrin Degradation Products as Screening Test for the
Diagnosis of Deep Vein Thrombosis. Haemostasis 1983; 13:254-261.
23. De Boer AC, Han P, Turpie AAG, Butt R, Zielinsky A, Genton E: Plasma and
urine Beta-Thromboglobulin concentration in patients with deep vein thrombosis
of the leg. Blood 1981; 58(4):693-698.
24. Van Hulsteyn H, Bertina R, Briet E: A one-year follow-up study of plasma
fibrinopeptide A and beta-thromboglobulin after deep vein thrombosis and
pulmonary embolism. Thromb res 1982; 27:225-229.
25. Koppen PW, Kuipers W, Hoegee-de Nobel E, Brommer EJP, Koopman J,
Nieuwenhuizen W: A quantitative enzyme immunoassay for primary
fibrinogenolysis products in plasma. Thromb Haemostas 1987; 57:25-28.
26. Elms MJ. Bunce JH, Bundesen PG, Rylatt DB, Webber AJ, Masci PP, Whitaker
AN: Measurement of crosslinked fibrin degradation products- An immunoassay
using monoclonal antibodies. Thromb Haemostas 1983; 50:591-594.
27. Rylatt DB, Blake AS, Glottis LE, Massingham DA, Fletcher WA, Masci PP,
Whitaker AM, Elms M, Bunce J, Webber A, Wyatt D, Bundesen PG: An
immunoassay for human D-dimer using monoclonal antibodies. Thromb Res
1983; 31:767-778.
28. Koppen PW, Hoegee-de Nobel E. Nieuwenhuizen WA: monoclonal antibody
based enzyme immunoassay for fibrin degradation products in plasma. Thromb
Haemostas 1988; 59:310-315.
29. Elms MJ, Bunce !A, Bundesen BG, Rylatt DB, Webber AJ, Masci PP, Whitaker
AN: Rapid detection of cross-linked fibrin degradation products in plasma using
monoclonal antibody-coated panicles. Am J Clin Path 1986; 85:360-364.
30. Koopman J, Haverkate F, Kopperl PW, Nieuwenhuizen W, Brommer EJP, van
der Werf WGC: New enzyme immunoassay of fibrin-fibrinogen degradation
products in plasma asing a monoclonal antibody. J Lab Clin Med 1987; 109:75-
91
84.
31. Graeff H, Hafter R: Detection and relevance of crosslinked fibrin derivatives in
blood. Sem Thromb Haemostas 1982; 8:57-68.
32. Van Bergen PFMM, Knot EAR, Jonker JJC, De Boer AC, De Maat MPM: Is
quantitative determination of fibrin(ogen) degradation products and thrombin
antithrombin Ill complexes useful to diagnose deep venous thrombosis in
outpatients? Thromb Haemost 1989; 62:1043-1045.
33. Hull R, VanAken WG, Hirsch J, Gallus AS, Hoicka G, Turpie AG, Walker I,
Gent M: Impedance plethysmography using the occlusive cufftechnigue in the
diagnosis of DVT. Circulation 1976; 53:696-700.
34. Amiral J. Plassart V, Minard F: Measurement and clinical relevance of D-dimer
by ELISA. ln:Fibrinogen and its derivatives. Muller-Berghaus G, Scheefers
Borchel U, Selmayr E, Henschen A (Eds.) Amsterdam-New York-Oxford:
Excerpta Medica, 1986, p285-290
35. Heaton DC, Billings JD, Hickton CM: Assessment of D dimer assay for the
diagnosis of deep vein thrombosis. J Lab Clin Med 1987; 110:588-591.
36. Bounameaux H, Schneider PA, Reber G, De Moerloose P, Krahenbuhl B:
Measurement of plasma D-dimer for Diagnosis of deep venous thrombosis. Am
J Clin Path 1989: 91:82-85.
37. Ott P, Astrup L, Jensen RH, Nyeland B, Pedersen B: Assessment of D-Dimer
in plasma: Diagnostic value in suspected deep venous thrombosis of the leg.
Acta Med Scand 1988: 224:263-267.
38. Rowbotham BJ, Carroll P, Whitaker, Bunce !A, Cobcroft RG, Elms MJ, Masci
PP, Bundesen PG, Rylatt DB, Webber AJ: Measurement of crosslinked fibrin
derivatives- use in the diagnosis of venous thrombosis. Thromb Haemostas
1987; 57(1):59-61.
39. Canova G, Cogo A, Cuppini S, De Toni R, Lensing AWA, Prandoni P:
Thrombin-antithrombin lll complexes m the diagnosis of deep venous
thrombosis. Fibrinolysis 1990; 4(1):21.
92
Chapter 4
FIBRIN AND FIBRINOGEN DEGRADATION PRODUCTS IN PLASMA
DURING GRAM-NEGATIVE SEPTIC SHOCK
H. Kroneman1•2
, E.S.J. Kalter, E.A.R. Knot', W. Nieuwenhuizen4
Department of Internal Medicine ll1, University Hospital Rotterdam-Dijkzigt, Center
for Clinical Decision Analysis2• Erasmus University Rotterdam, Department of Intensive
Care3• University Hospital Nijmegen.Current address: Centocor B.Y .. Department of
Clinical Research. Leiden, Gaubius Laboratory-IYYO-TN04, Leiden,'The Netherlands
Submitted for publication
ABSTRACT
Gram-negative septic shock is an increasingly common event associated with a high
mortality. Complicating factors, such as multiple organ failure and adult respiratory
distress syndrome are due to widespread fibrin depositions, and are considered to
depend largely upon extreme activation of the coagulation pathway. It has recently been
suggested, however, that impairment of fibrinolysis may also play a role. In this study
we foccussed on the fibrinolytic system in 30 patients with septic shock, by measuring
plasma levels of the total fibrin and fibrinogen degradation products (TDP) and
fibrinogen degradation products (FgDP). Mean TDP and FgDP levels were elevated in
patients with septic shock and showed a tendency toward normalization during follow
up. The mortality rate was 50%. No difference could be detected between plasma TDP
and FgDP levels in survivors and nonsurvivors of septic shock. This study suggests that
the fibrinolytic system is activated during septic shock. Further studies are needed to
determine whether activation of the fibrinolytic system in patients with septic shock
may improve the prognosi;;.
94
INTRODUCTION
Gram-negative sepsis is an increasingly common clinical event with a mortality rate
between 20% and 50%. 12 In patients who develop septic shock. mortality exceeds 40%
and may even approach 90% in patients with serious underlying disease.'
Endotoxins, lipopolysaccharide, and especially the lipid-A fraction of the gram-negative
bacterial cell wall, trigger the pathophysiologic response to gram-negative septicemia!
They exert numerous biological effects, such as generation of vasoactive peptides.S·' and
activation of coagulation' and complement pathways.' Endotoxin is also a potent
stimulator for the release of tumor necrosis factor (TNFa) from macrophages and
monocytes! TNF activates factor X and thus the common coagulation pathway. 10
Septicemia is frequently associated with disturbances of the hemostatic balance.11
During septic shock, decreased levels have been detected of prekallikrein and
coagulation factors XII, XI and Vll, suggestive for consumption of these factors. 12.1'
The activation of the coagulation system may cause disseminated intravascular
coagulation (DIC) with widespread depositions of fibrin in the microvasculature and
subsequent development of multiple organ failure. 14.I
5 The activation of kallikrein in
turn m<>y also cause the liberation of bradykinin and the activation of plasmin.
Bertozzi and coworkers recently demonstrated that inhibition of fibrinolytic activity due
to an increase in urokinase inhibitors and amiplasmins contributes to the formation and
persistence of hyaline membranes in the lung and thus to the development of adult
respiratory distress syndrome. a common complication of septic shock.16 Furthermore
it has been demonstrated that endotoxin administration to normal subjects results in
early activation of fibrinolysis through release of tissue plasminogen activator (t-PA),
which is abruptly terminated by a rise of plasminogen activator inhibitor I (PAl-l)
activity." Extremely high PAl-l levels have indeed been encountered in patients with
septicaemia12.1 8
•19 suggesting that plasmin mediated lysis of fibrin and fibrinogen may
be impaired during septic 'hock.
Fibrinolysis is an important mechanism in protecting the circulation from massive
thrombosis and is reflected by the formation of degradation products of fibrin (FbDP)
and fibrinogen (FgDP). Until recently, blood levels of fibrin and fibrinogen
95
degradation products could not be detected reliably, because they could only be
assessed in serum. The use of serum samples leads to unreliable results?021
During the past few years, enzyme immuno assays (EIA's) have become available for
quantitative assessment of fibrin and fibrinogen degradation products in plasma. The
aim of our study was to investigate the fibrinolytic state in patients with septic shock
by measuring degradation products of fibrinogen (FgDP) and the total of fibrin and
fibrinogen degradation products (TDP) in plasma.
PATIENTS AND METHODS
This investigation was performed as an ancillary study to the so called Swiss-Dutch
nial, which was designed to determine the efficacy of hyperimmune lgG against E-coli
J5 in comparison to a normal lgG preparation in patients with gram-negative septic
shock.22 Blood samples were obtained from 30 consecutive patients in the Netherlands
(participating centres and principal investigators: Sint Radbout Hospital Nijmegen (E.
Kalter), University Hospital Utrecht (J. Schellekens and J. Verhoef), who fulfilled the
inclusion criteria of the Swiss-Dutch trial: L The presence of a gram-negative
bacteremia, a proven or firmly suspected gram-negative focal infection or a clinical
condition predisposing to gram-negative infection (e.g. granulocytopenia, septic
abdominal surgery, esophageal surgery or surgery for ruptured aonic aneurysmal and
2. the presence of septic shock based on clinical grounds and at least one of the
following signs: oliguria (<30 ml/h), hypothermia (<36°C), hypoxentia (P02<30 mm
Hg), respiratory alkalosis (pH >7.5 or Pc02<30 mm Hg), metabolic acidosis (pH <7.3
or base excess <-10 mmol/1), coagulation abnormalities (plasma thromboplastin level
of <50% of normal, APIT >40s, or the elevation of fibrin split products in serum) and
thrombocytopenia (<I 00 OOO/mm3 or a decrease of>50% of a previously normal value).
All patients received standard therapy for septic shock (fluid resuscitation, vasopressors
and antibiotic treatment) •nd support of the vital functions if required (Swan-Ganz
catheterization, artificial ventilation and hemodialysis). In addition, the patients were
randomized to receive a single intravenous dose of 200 mg/kg (max 12 g) of either a
standard lgG preparation or a preparation of human lgG (J5) antibody to Escherichia
96
coli during one hour.
Blood samples were taken prior to infusion (t=O), 2 hours after infusion (t=l ), 24 hours
after infusion (t=2) and after 10 days (t=3). Plasma was obtained by cenrrifuging a
mixture of nine volumes of blood and one volume of 0.11 M sodium citrate for 30
minutes at 2000 x g at 4' C. The plasma samples were stored frozen in small aliquots
at -20'C until tested and carefully thawed at 37'C prior to assay. In order to be able to
establish normal reference values, venous blood samples were obtained from 43 non
patient volunteers (male:female = 24:19: median age: 31 yrs, range: 22-45 yrs) and
processed in an identical manner.
Assays
FgDP and TOP were determined in plasma using specific EIA's of the sandwich type
based on monoclonal antibodies (Organon Teknika, Turnhout, Belgium). The capture
antibody was FDP-14, the tagging antibody for detection in the FgDP EIA was FDP
Y-18.23 A combination ofFDP Y-18 and FDP DD-13 was used as tagging antibody in
the EIA for TOP assessment.''
Statistical analysis
Results are given as mean ± standard error of the mean (s.e.m.), unless reported
otherwise. Differences between groups were tested by applying the Wilcoxon ranksum
test for independent samples. Within group differences were determined by Wilcoxon
matched-pairs signed rank test. Any P value <0.05 was considered to represent a
significant difference. All reported significance levels were two-sided. Regression
analysis was performed by method of least squares and the correlation coefficient was
calculated.
RESULTS The mortality rates were 50% in either of the treatment groups. We therefore analysed
all data assuming that treatment difference did not exist.
Of the 30 patients with a gram-negative septic shock (median age:64, range 21-78,
male/female ratio: 20:10), 15 died due to septic shock; 8 of whom died within 7 days.
The median plasma values of the TDP and FgDP in normal healthy controls and
97
patients with septic shock are shown in table l.
The mean plasma TOP concentration in the baseline plasma sample (t=O) of patients
with septic shock (6.52 ug/ml ± 1.83) was markedly elevated (p<O.OOOI) as compared
with that of healthy subjects (0.42 ug/ml ± 0.02). In addition, patients with septic shock
had significantly higher plasma FgOP levels (2.59 ug/ml ± 0.93) than healthy subjects
(0.20 ug/ml ± 0.004, p<0.0001).
Patients TDP FgDP
normals 0.42 (0.21-0.65) 0.19 (0.17-0.29) sepsis 3.01 (0.25-42.9) 0.65 (0.25-19.9) survivors 2.61 (0.25-42.9) 0.62 (0.25-19.4) nonsurvivors 3.09 (0.25-36.5) 0.81 (0.25-19.9) death <7 days 3.09 (0.25-36.5) 0.43 (0.25-19.9) death >7 days 2.90 (2.29-14.4) 1.01 (0.25-5.50)
Table I Median and mnge of TDP and FgDP plasma levels (ug/ml) in patients with
gram-negative septic shock and normal subjects.
The differences between survivors and nonsurvivors of septic shock at t=O were not
significant for mean TOP (6.65 ± 2.89 vs 6.38 ± 2.35 respectively. p=0.53) and mean
FgDP (2.65 ± 1.37 vs 2.47 ± 1.29 respectively, p=0.72) plasma levels.
The mean ratio of FgDP!fDP in patients with septic shock (0.39 ± 0.05 ug/ml) in the
first plasma sample was lower (p<O.Ol) than in healthy subjects (0.52 ± 0.02 ug/ml).
The difference in mean FgDP(fOP ratios between survivors (0.46 ± 0.08 ug/ml) and
nonsurvivors (0.34 ± 0.06 ug/ml) was not statistically significant (p=0.12).
Fig l shows the mean plasma levels of TOP. FgOP and the FgOP!fDP ratio of patients
with septic shock in the first plasma sample, after 3 hours, after 24 hours and after 10
days. The variation in mean TOP and FgOP values was not significant (0.32<p<0.86
98
and 0.22 <p<0.76 respectively). The mean FgDP{fDP ratio remained unchanged
(p=0.03) after 3 hours (0.44 ± 0.06 ug/ml). but declined significantly (p<0.005) to 0.29
± 0.04 ug/ml after 24 hours. During follow-up. no significant differences were observed
in mean TDP plasma levels (0.39<p<0.71) and FgDP values (0.35<p<0.72) between
survivors and nonsurvivors.
FgDP values correlated extremely well with TDP values in the first plasma sample
(r=0.97, p<O.OOOI, n=30) and after 3 hours (r=0.96, p<O.OOOI, n=29). However, this
correlation disappeared after 24 hours (r=0.42, p<0.05, n=24) and after 10 days (r=0.33,
p=0.18, n= 18).
10
9
~ 8 2 ~ 7 0 >-
~ 6 u.
~ 5 • :a u. 4
0: 3 0
>-c • 2 ~
0
Figure 1.
- TOP
FgDP
FgDP/TDP
3 hours 24 hours I 0 days
Mean levels of TDP and FgDP and the ratio of FgDP{TDP in patients with
septic shock !n the first plasma sample. after 3 hours, 24 hours and and 10
days.
DISCUSSION
In this study we investigated the fibrinolytic system in patients with septic shock by
measuring FgDP and TLP plasma levels with new sensitive EIA's. One of the
99
pathophysiological hallma:-ks of septic shock are the DlC related events, such as fibrin
deposition at the microvascular level1 Degradation of fibrin, deposited in the
microvasculature, by a physiological fibrinolytic response is an important mechanism
of the organism in the defence against massive thrombosis.
The fibrinolytic activity depends primarily on the balance between t-PA and PAl-l
activity. Recent studies suggest that the fibrinolytic system may be impaired in
septicemia through an increased PAl-l activity.18•19 High PAI-I levels appear to be
associated with a poor prognosis of septic shock.12-'
9 Although measurement of both t
PA and PAI-l seems thus of importance in the prognosis of septic shock, circulating
levels of fibrin and fibrinogen degradation products could be a more reliable measure
of their resultant activity in blood. The most tangible evidence for ongoing fibrinolysis
in blood is the presence of degradation products of fibrin and fibrinogen. Plasma levels
of TDP and FgDP were markedly elevated in patients with septic shock as compared
with healthy subjects. This is in agreement with findings of others who found elevated
D-dimer levels during septic shock 1225 or increased FgDP and TDP levels in patients
with septicemia20 and suggests that simultaneous activation of both the coagulation and
the fibrinolytic system occur in these patients. Moreover, the elevated levels of FgDP
indicate the occurrence of primary fibrinolysis (fibrinogenolysis), which has been shown
to accompany fibrinolysis in many cases27. FgDP's possess anticoagulant properties29
'31
and their presence in plasma could theoretically helping preventing the formation of
microthrombi. However, this study suggests that their contribution to the course of
septic shock is limited as no difference could be detected between mean FgDP levels
of survivors and nonsurvivors. It could well be that their contribution is of greater
importance at an earlier stage.
No difference could be observed between survivors and nonsurvivors in mean TDP and
FgDP plasma levels in the. first plasma sample. Measurements during the follow-up
period of the whole group showed that there was a tendency to normalization in plasma
levels of degradation products (see Fig.!). This tendency was slightly more pronounced
in survivors than in nonsurvivors. but the difference was not significant. Whether this
latter is due to a type ll error needs to be determined by studying larger groups. The
100
decline of the FgDP[fDP ratio in the sample after 24 hours may be explained by an
increase of FbDP plasma levels, since TDP is the sum of FgDP and FbDP.28
In conclusion, this study shows that TDP and FgDP levels are significantly elevated
during septic shock, indicating the process of ongoing fibrinolysis. No difference could
observed between levels of degradation products of survivors and nonsurvivors, which
suggest that the fibrinolytic activity was comparable in both groups. This study was not
designed to determine whether the extent of fibrinolytic response is adequate for the
degradation of fibrin deposits and microvascular thrombi during septic shock. Further
studies are needed to see whether stimulation of fibrinolytic activity. e.g. by preventing
the increase of PAI-1 or administration of t-PA. has a beneficial effect on the prognosis
of septic shock. That this may be the case is suggested by experimental results of t-PA
therapy on glomerular thrombi in the Shwanzman reaction in rabbits?2
ACKNOWLEDGMENT
The authors are much indebted to Th. Calandra, M.P. Glauser (Centre Hospitalier
Universitaire Vaudois. Lausanne, Switzerland), J. Schellekens and J. Verhoef
(University Hospital Utrecht) for providing the excellent background of this ancillary
study and inherently to all other participants, especially those supporting the Dutch arm
of the study: Prof. J.M.C. Douze. Dr. P.F. Hulstaert. Dr. L.E.C. Verhoeven-Van Duin
(Utrecht) and Dr. J.S.F. G<mbrere (Nijmegen).
REFERENCES
Young LS. Gram-negative sepsis. In: Mandell GL, Douglas RG Jr. Bennett JE,
eds. Principles and practise of infectious diseases. 3rd ed. New York: Churchill
Livingstone; 1990: 611-36.
2 Centers for disease .oontrol. Increase in national hospital discharge survey rates
for septicemia-United States, 1979-1987. Jama !990; 236: 937-938.
3 Bone RC, Fisher CJ Jr, Clemmer TP et al. A controlled clinical trial of high
dose methylprednisolone in the treatment of severe sepsis and septic shock. N
Engl J Med 1987; 317: 653-658.
101
4. Ryan RL. Microbial factors in pathogenesis: Lipopolysaccharides. ln: Root RK,
Sande MA, eds. Septic Shock. New York, Churchill Livingstone 1985: 133-25.
5. Kimball HR. Melmon KL, Wolff SM. Endotoxin induced kinin production in
man. Proc Soc Exp Bioi Med 1972; 132: 139
6. Brandtzaeg P, Oktedalen 0, Kierulf P, Opstad PK. Elevated VIP and endotoxin
plasma levels in human gram-negative septic shock. Regulatory peptides 1989;
24: 37-44
7. Levy GA, Schwartz BS, Curtiss LK, Edgington TS. Regulatory roles of T-mu
and T-gamma cells in the collaborative cellular initiation of the extrinsic
coagulation pathway by bacterial lipopolysaccharide. J Clin Invest 1985; 76:
548-555.
8. Vukajlovitch SW. Hoffman J, Morrison DC. Activation of human serum
complement by bacterial Iipopolysaccharides: structural requirements for
antibody dependent activation of the classical and alternative pathways. Mol
Immunol 1987; 24: 319-332.
9. Beutler B. Mahoney J. LeTrang N. Pekala P. Cararni A. Purification of
cachectin, a lipoprotein lipase-suppressing hormone secreted by endotoxin
induced RAW 264.7 cells. J Exp Med 1985; 161: 984-995.
10. Poll van der P, Buller HR, Ten Cate H. Wortel CH, Bauer KA, VanDeventer
SJH. Hack E, Sauerwein HP, Rosenberg RD, Ten Cate JW. Activation of
coagulation after administration of tumor necrosis factor to normal subjects. N
Eng! J Med 1990; 322: 1622-1627.
II. Smith-Erichsen N, Aasen AO. Gallimore MJ eta!. Studies of components of the
coagulation system in normal individuals and septic shock patients. Circ Shock
1982; 9:491
12. Hesselvik JF. Blomback M. Brodin B. Mailer R. Coagulation, fibrinolysis, and
kallikrein systems i~ sepsis: relation to outcome. Crit Care Med !989; 17: 724-
733.
13. Kalter ES, Daha MR, Ten Cate JW, Verhoef J, Bouma BN. Activation and
inhibition of Hageman factor-dependent pathways and the complement system
102
in uncomplicated bacteremia or bacterial shock. J Infect Dis 1985; 151: 1019-
1027.
14. Marder VJ. Martin SE. Francis CW. Colman RW. Consumptive
thrombohemorrhagic disorders. In: Colman RW, Hirsch J, Marder VJ, Salzman
EW, eds. Hemostasis and Thrombosis: basic principles and clinical practise. 2nd
ed. Philadelphia:J.B. Lippincott, !987: 975-1015.
15. Carrico CJ, Meakin> JL, Marshall JC, Fry D, Maier RV. Multiple-organ-failure
syndrome. Archives of Surgery 1986: 121: 196-208.
16. Benozzi P, Astedt B, Zensius L, Lynch K, Lemaire F, Zapol W, Chapman HA
Jr. Depressed bronchoalveolar urokinase activity in patients with adult
respiratory distress syndrome. N Eng! 1 Med 1990: 322: 890-897.
17. Suffredini AF. Harrel PC. Parrillo JE. Promotion and subsequent inhibition of
plasminogen activation after administration of intravenous endotoxin to normal
subjects. N Eng! J Med 1989: 320: 1165-1172.
18. Engebretsen LF. Kierulf P. Brandtzaeg P. Extreme plasminogen activator
inhibitor and endotoxin values in patients with meningococcal disease. Thromb
Res 1986; 42:713-716.
19. Pralong G, Calandra T. Glauser MP, Schel!ekens J, Verhoef J, Bachmann F,
Kruithof EKO. Plasminogen activator inhibitor 1: a new prognostic marker in
septic shock. Thromb Haemostas 1989; 61: 459-462
20. Nieuwenhuizen W. Plasma assays for derivatives of fibrin and fibrinogen, based
on monoclonal antibodies. Fibrinolysis 1988; 2: 1-5.
21. Gaffney PJ, Perry MJ. Unreliability of current serum fibrin degradation products
(FDP) assays. Thromb Haemostas 1985; 53: 301-302.
22. Calandra T. Glauser MP, Schel!ekens J, Verhoef J. the Swiss-Dutch J5
immunoglobulin study group. Treatment of gram-negative septic shock with
human lgG antibody to Escherichia Coli J%: a prospective, double-blind,
randomized triaL J Infect Dis 1988; 158: 312-319.
23. Kopper! PW. Kuipers W. Hoegee-de Nobel E, Brommer EJP, Koopman J,
Nieuwenhuizen W. A quantitative enzyme immunoassay for primary
103
fibrinogenolysis products in plasma. Thromb Haemostas 1987; 57: 25-28.
24. Koopman J. Haverkate F, Koppen PW, Nieuwenhuizen W, Brommer EJP, van
der Werf WGC. New enzyme immunoassay of fibrin-fibrinogen degradation
products in plasma using a monoclonal antibody. J Lab Clin Med 1987; 109: 75-
84.
25. Voss R , Matthias FR, Borkowski, Reitz D. Activation and inhibition of
fibrinolysis in septic patients in an intensive care unit. Br J Haematol 1 990; 75:
99-105
26. Takahashi H, Tatewaki W, Wada K, Niwano H, Shibata A. Fibrinolysis and
fibrinogenolysis in disseminated intravascular coagulation. Thromb Haemostas
1990; 63: 340-344
27. Nieuwenhuizen W. A double blind comparative study of six monoclonal
antibody-based plasma assays for fibrinogen derivatives. In: Lowa GDO,
Douglas JT, Forbes CD, Henschen A, eds. Fibrinogen 2, biochemistry,
physiology and clinical relevance. Amsterdam, Excerpta Medica, 1987; 181-186.
28. H. Kroneman. W. Nieuwenhuizen, EAR Knot. PFMM van Bergen, MPM de
Maat. Correlations between plasma levels of fibrin(ogen) derivatives as
quantified by different assays based on monoclonal antibodies. Thromb Res
1991; 4: 441-452.
29. Nieuwenhuizen W. Voskuilen M, Hermans]. Anticoagulant and calcium-binding
propenies of high molecular weight derivatives of human fibrinogen (plasmin
fragment Y). Biochim Biophys Acta 1982; 708:313-316.
30. Haverkate F. Timan G. Nieuwenhuizen W. Anticlotting properties of fragments
D from human fibrinogen and fibrin. Eur J Clin Invest 1979; 9: 235-255.
31. Marder VJ, Schulman NR. High molecular weight derivatives of human
fibrinogen produced by plasmin. ll. Mechanism of their anticoagulant activity.
J Bioi Chern 1969; 244: 2120-2144
32. Bergstein JM, Riley M. Tissue plasminogen activator therapy of glomerular
thrombi in the Shwartzman reaction. Kidney lnt 1989; 35: 14-18.
104
Chapter 5
PHARMACOKINETICS OF LOW MOLECULAR WEIGHT HEPARIN AND
UNFRACTIONATED HEPARIN DURING ELECTIVE AORTOBIFEMORAL
BYPASS GRAFTING
H. Kroneman1·', B.C. Eikelboom5
, E.A.R. Knot\ P. de Smir, T.H.N. Groenland3,
M.P.M. de Maat\ H. Van Ur~.
Depts. of Internal Medicine 1!1, Surgery" and Anesthesiology', University Hospital
Rotterdam-Dijkzigt, Center for Clinical Decision Analysis', Erasmus University
Rotterdam. Dept. of Surgery5, St. Antonius Hospital, Nieuwegein, The Netherlands
J Vase Surg 1991;14:208-14
ABSTRACT
Peroperative monitoring h~s demonstrated that administering of heparin on empirical
basis is associated with a wide variation in patient response and elimination rate. This
problem may be overcome by intervention on the basis of peroperative monitoring or
by using heparins with different pharmacokinetic properties. When compared with
unfractionated heparin (UFH), low molecular weight heparins (LMWH) have a higher
bioavailability after subcutaneous administration, a linear clearance mechanism with a
prolonged half life and are at least as effective in preventing postoperative vein
thrombosis. Theoretically these characteristics ofLMWH could lead to more predictable
levels of heparin activity.
In this study we compared the pharmacokinetics of LMWH and UFH after an
intravenous injection in p2tients undergoing aortic graft surgery. Heparin activity was
measured before heparin administration and at 5, 20, 35, 50, 65, 80, 95 and l!O
minutes after administration. The anti-Xa activity in the LMWH group showed less
variation and was more sustained when compared to the UFH group. Fibrin degradation
products were moderately correlated with the anti-factor Xa levels of the LMWH
group, but no correlation was found in the UFH group. The anti-factor Xa activity of
LMWH was, in contrast to that of UFH. not completely reversible by protamine
administration. The blood loss was comparable in both groups. In contrast to what was
expected, the pharmacokinetic profiles of LMWH and UFH showed a similarity after
intravenous injection in patients undergoing aortobifemoral bypass grafting. Factors that
could have influenced the pharmacokinetic behaviour of the heparins are discussed.
106
INTRODUCTION
The use of heparin in the surgical treatment of arteries has first been described in 1940
and it was shown that heparin had a favourable effect on vascular patency'. Gradually,
the administration of heparin during reconstructive arterial surgery has become widely
accepted for the prevention of intravascular peroperative thrombosis. At present, the
majority of surgeons routinely anticoagulate their patients by giving a fixed dose
heparin (mostly 4000-8000 IU) intravenously'·'. although some prefer a bodyweight
dependent dose of heparin i.v-'·10• Previously, we have studied heparin levels during
abdominal aortic reconstructions4. In accordance with other studies2
•10
-13
, it was
demonstrated that patients response to heparin showed a large interindividual variability,
which makes it difficult to maintain predictable or adequate levels of anticoagulation
for all patients. One way to attack this problem may be the close monitoring of heparin
levels in each patient. another approach is to study the utility of heparins with different
pharmacological properties.
The heterogeneous nature of unfractionated heparin (UFH), like its polydispersity of
molecular weight and different chemical properties. may contribute to the variable
response in patients. After intravenous administration. the disappearance curve of UFH
anticoagulant activity. as r.Jeasured with the APTT or with the anti-Xa assay, follows
non-linear kinetics and is characterized by a rapid initial disappearance phase followed
by a convex curve". During aortobifemoral bypass grafting. the duration of the aortic
clamping rime is such. that the UFH elimination will mainly occur in this rapid
disappearance phase.
In the search for improvement of heparin therapy, attempts have been made to develop
heparin derivatives with modified pharmacological and biological properties. One group
of heparin derivatives, the low molecular weight heparins (LMWH), have become of
great clinical interest. Some LMWH preparations have already been approved and
registered for clinical use. LMWH are derivatives of UFH characterized by a shorter
oligosaccharide chain leng,h and a prolonged half life time in comparison to the parent
heparin. LMWH' s exert anticoagulant activity in a different way than UFH. A
minimum of 18 monosaccharide units is needed for the simultaneous binding of
107
thrombin and antithrombin on a heparin template, while the saccharide size needed for
interaction with factor Xa is smaller15• Hence, because of their shorter chain length,
LMWH's have a diminished anti-factor lla activity, which is reflected by a limited
ability to prolong clotting time, but are potent enhancers of the inhibition of coagulation
factor Xa.
Following subcutaneous administration, the bioavailability of LMWH is high (± 90%)
in comparison with UFH (± 15%)16, which makes it possible to reach therapeutical
levels in thrombotic patients without dose adjustments17• Analysis of randomized
controlled level! trials comparing UFH and LMWH in the prophylaxis of postoperative
venous thrombosis demonstrated a clinically important difference in efficacy in favour
of LMWW8 After intravenous administration LMViH's follow a linear elimination
pattern19.21 Theoretically, the altered pharmacokinetic properties of LMWH's could
result in more predictable levels of heparin activity during aortic reconstructive surgery.
The purpose of this study was to compare the pharmacokinetics of intravenously
administered LMWH and UFH in patients undergoing elective aonobifemoral bypass
surgery.
PATIENTS AND METHODS
patients
The study included consecutive patients undergoing an elective aonobifemoral bypass
prosthesis operation. Patients were excluded from the study if they had (previous)
bleeding disorders, if they had taken anticoagulant or anti platelet therapy during 5 days
prior to surgery or if they had not given informed consent. All exclusions were
documented.
A knitted, gelatin impregnated Dacron graft ( Yascutek', lnchinnan, Renfrewshire,
Scotland) was used for bypass surgery. Standardized techniques of aortic graft insertion
were used in all procedures. The proximal anastomosis to the aortic neck was made
either end-to-end when it concerned aortic aneurysmal or end-to-side in case of
aonoiliac occlusive disease. The distal pans of the graft were sutured end-to-side to the
common femoral arteries just proximal of the bifurcations of the superficial and deep
108
femoral arteries. Flushing of the graft just before completion of the anastomoses was
also part of the procedure.
Study design
The study was performed as a randomized study to compare the pharmacokinetics of
LMWH and UFH administered during elective abdominal aortobifemoral prosthesis
operations. Included patients were randomly assigned to receive LMWH (dosage: 90
anti-Xa units/kg) or unfractionated heparin (dosage: 80 anti-Xa units/kg). The operating
surgeons were kept unaware of the assignment. Heparin was administered as a bolus
injection intravenously 3 minutes prior to clamping of the aorta.
Apart of the heparinization, no further attempt was made to alter the surgeon· s existing
empirical protocol. A solution containing 25 anti-Xa IU UFH per ml normal saline was
used for flushing. Protamine was administered if and as desired by the operating
surgeon after completion of the anastomosis and removal of the clamps and before
abdominal closure. Blood samples were taken after induction of anaesthesia (before
heparin administration). 5 minutes after heparin administration and each 15 minutes
thereafter until completion of the anastomosis and removal of the aorta clamp. If
protamine was administered, a blood sample was taken 5 minutes later.
The study was approved by the medical ethical committees of the University Hospital
Rotterdam - Dijkzigt. Rotterdam and the St. Antonius Hospital, Nieuwegein, the
Netherlands. Informed consent was obtained from each eligible patient before entering
into the study. The study was carried out according to the declaration of Helsinki.
Heparins
- UFH: Thromboliquine' (Organon, Oss, the Netherlands) in prepacked syringes of 5
ml containing 5.000 IU/ml of the highly purified sodium salt of heparin from animal
source.
- LMWH: the used LMWH (Kabi 2165, Fragmin', Kabivitrum AB, Stockholm,
Sweden) was obtained by partial depolymerization of heparin from porcine intestinal
mucosa with nitrous acid and isolated by ethanol precipitation. Fragminr has an average
109
MW of 4000-6000 Daltons and about 90% of the material between MW 2000-9000.
The specific anticoagulant activities are about 160 U/mg in an anti-factor Xa assay
using the chromogenic substrate S-2222 (Kabivitrum, Stockholm) and about 40 U/mg
in the APTT and thrombin inhibitions assays. Aqueous solutions of Fragmin' were
supplied in prepacked syringes containing 10.000 IU in a 1 ml solution.
Laboratory assays
- Assays were performed using the FP-910 coagulation system (Labsystems, Helsinki,
Finland) as described earlier'.
- Heparin activity was assessed by measuring anti-factor Xa activity using the
chromogenic substrate S-2222, bovine factor Xa and human AT-Ill concentrate (Coatest
heparin. Kabi Vitrum, Amsterdam, the Netherlands )22 A standard curve was constructed
by testing normal citrated plasma samples (obtained from apparently healthy hospital
employees) containing increasing amounts of heparin at concentraripns ranging from
0-1 U/ml.
-Activated partial thromboplastin time (AP1T) was determined with actin (Merz Dade,
Dijdingen, Switzerland)".
-Fibrin degradation products (FbDP's) were quantitatively determined in plasma using
an enzyme immunoassay (E!A) of the sandwich type (Organon Teknika, Turnhout,
Belgium). This E!A is based upon a monoclonal antibody as capture antibody (FDP-
14). The tagging antibody was a horse-radish peroxidase (HRP) conjugated monoclonal
antibody (DD-13) against the D-dimer''.
- Platelet count. hemoglobin and hematocrit were determined in EDTA-blood.
Thrombotest (TT) was also determined.
Blood samples
Venous blood samples were collected from an arterial line into 5 ml tubes containing
3.2% trisodium citrate (9:1). The first 10 ml of blood was discarded to avoid
contamination of the sample with heparin used for flushing. The blood samples were
immediately cooled on melting ice and were centrifuged in an Eppendorl centrifuge
(3.5 min, 10.000 g) or a Hettich centrifuge (10 min, 3200 gat 4'C). Plasma samples
110
were stored frozen in smdl aliquots at -70°C and carefully thawed at 37°C prior to
assay.
Blood loss
The intraoperative blood loss was accurately collected and recorded through suction and
gauzes.
Statistical analysis
Results are given as mean ± standard deviation unless reported otherwise. Within
groups differences between the means were tested using Wilcoxon's signed Rank test
for paired samples. Differences between the two groups were tested with the two
sample Wilcoxon ranksum test. All tests were two-sided and any probability less than
5% was considered statist.ically significant. Pearsons correlation test was used for
comparison of the APTT and the anti-Xa levels and for comparing the FbDP levels
with the anti-Xa levels.
RESULTS
Patients
Fifty three consecutive patients underwent elective abdominal aortobifemoral bypass
grafting. Thirteen patients were excluded from the study. one because a history of
excessive bleeding during previous surgery, eight because they had been taking
anticoagulant or antiplatelet drugs within 5 days prior to surgery and four because they
were unwiHing to give inf0rmed consent.
Forty eligible patients who were willing to give informed consent entered into the
study. Twenty patients were assigned to receive UFH and twenty to receive LMWH.
The patients characteristics are shown in table I.
The indications for operation were aortoiliac occlusive or aneurysmal disease of the
aorta. Fifteen patients (38%) suffered from aortoiliac occlusive disease (7 of the UFH
group and 8 of the LMWH group). Twenty-five patients were electively operated upon
because of an aortic aneurysma (!3 of the UFH and 12 of the LMWH group). The
aorta was cross-clamped above the renal arteries (suprarenal) in 15% of the cases (4
patients in the UFH group and 2 in the LMWH group). The preoperative hematologic
111
Bodyweight age male/female
Group (Kg) (yrs) ratio
UFH 75.7 ± 16.1 61.2 ± 11.1 17/3 LMWH 71.6 ± 8.5 64.9 ± 9.7 18/2 Total
Table 1
73.6 ± 12.7 63.0 ± 10.3 35/5
Characteristics of patients undergoing elective aortobifemoral bypass
grafting (mean ± std).
UFH LMWH
Platelet count 222 ± 67 233 ±57
(x 109 /l)
Haemoglobin 8.6 ±1.1 8.9 ± 0.8 (mmol/1)
haematocrit 0.40 ± 0.05 0.42 ± 0.04 Thromborest 74% ± !7% 65% ± 23% APTT (sec) 38.3 ± 12.4 34.8 ± 6.06
Table 2 Preoperative hematologic data of patients receiving UFH or LMWH.
Group
UFH LMWH
Total
Table 3
Blood loss Operation time Clamping time
mean s.e.m. mean s.e.m. mean s.e.m.
242! 296 235 2! 72 !0 2098 438 207 13 6! 6 2255 357 22! 12 66 6
Intraoperative blood loss (ml), duration of operation (min)and clamping time
of the aorta (min) in aortobifemoral bypass grafting. Results are given as
mean and standard error of the mean (s.e.m.).
l12
data of both groups were comparable (see table 2). The intraoperative blood loss, the
operation time and the clamping time of the aorta in both groups are shown in table 3.
The between groups differences of the means were statistically not significant.
Heparin levels
The disappearance curves of the two heparins, as assessed by the levels of anti-factor
Xa activity, are shown in figure I.
c > :;:; 0
"' "' X I
c "'
Figure I
2,------------------------------,
UFH mean
LMWH mean
QL---~--~--L-~~~--~--~--~---"
BH 5 20 35 50 65 80 95 1 1 0 125 time after hepann adm1n1stration (min)
The ami-factor Xa activity (IU/ml) before (BH) and after heparin
administration in patients receiving UFH or LMWH (mean ± SD). The arrow
indicates an increase of UFH concentrations following flushing of the aona
with UFH preparations.
No anti-Xa activity was measurable in any of the samples before administration of
heparin. Five minutes after heparin infusion, the mean heparin level in the UFH group
113
and the LMWH group were 0.88 ± 0.43 anti-Xa U/ml and 1.06 ± 0.23 anti-Xa U/ml
respectively. This difference was not significant.
After removal of the clamps and establishment of flow the mean anti-Xa activity in the
LMWH group (0.30 ± 0.14 U/ml) was significantly higher (p<0.05) when compared
with the UFH group (0.21 ± 0.27 U/ml). The anti-Xa activity (mean± SD, in U/ml)
of the samples drawn at 20, 35, 50, 65, 80 and 95 minutes .after heparin administration
were: 0.67 ± 0.29 and 0.77 ± 0.18: 0.53 ± 0.28 and 0.59 ± 0.18: 0.44 ± 0.27 and 0.49
± 0.17; 0.38 ± 0.24 and 0.41 ± 0.17 ;0.40 ± 0.29 and 0.31 ± 0.15; 0.24 ± 0.25 and 0.32
0.09 ± 0.09 for UFH and LMWH respectively.
The anti-Xa levels (mean ± std) at 110 minutes after heparin administration were
significantly higher (p<0.05) in the LMWH group (0.25 ± 0.11 U/ml) when compared
with the corresponding values in the UFH group (0.05 ± 0.09 U/ml).
APTT measurements
UFH group
The intraoperative APTT values in the samples drawn before heparin administration
were 51.7 ± 18.2 seconds. Five minutes after heparin administration the APTT was
more than 300 seconds (which was the maximal detectable time in our assay) in 14
patients. Six patients had APTT levels ranging from 129 to 160 seconds. After
establishment of flow. 9 p&tients still had APTT levels above 300 seconds. Because the
maximum measurable AP1T value in our assay was 300 seconds, it was not possible
to calculate Pearsons correlation coefficients for comparing the APTT values with the
anti-factor Xa activity.
LMWH group
The mean APTT intraoperatively before heparin administration was 51.6 ± 18.4
seconds. The mean APTT value 5 minutes after heparin administration was 156 ± 76
seconds. which is significantly lower (p<O.OOl) when compared with the corresponding
values in the UFH group. None of the patients had APTT levels above the maximal
detectable level of 300 seconds. After removal of the clamp and establishment of flow
the APTT was significantly (P<0.05) diminished to 124 ± 46 seconds. No correlation
H4
was observed between the APTT levels and the anti-Xa activity (r=0,22; p=ns).
Fibrin degradation products
The median FbDP levels of the samples drawn before heparin administration were
0.630 ug/ml (range: 0.15-2,3 ug/ml) in the UFH group and 0.798 ug/ml (range: 0.15-
13.2 ug/ml) in the LMWH group. As can be seen in figure 2, the mean FbDP levels
increased significantly during operation in both groups (UFH group: p<O.Ol; LMWH
group: p<0.05). After completion of the anastomosis and removal of the clamps, the
median and range (in ug/ml) of the FbDP values were 0.956 (0.10- 3.46) in the UFH
group and !.53 (0.26 - 28.0) in the LMWH group. The between groups differences of
the FBDP values were not significant. Pearsons correlation coefficient for the FbDP
levels and the anti-Xa activity was 033 (p<0.05) in the LMWH group. No correlation
was found (r=O.Ol4, p=ns) for FbDP levels and anti-Xa activity in the UFH group.
10
9
8
7
6 E
" 5 ~
4
3
2
0
Figure 2
before heparin end operation
111111111111 UFH Mean FbDP
C:=J LMWH Mean FbDP
Intraoperative FbDP plasma levels {ug/ml) before heparin administration and
at the end of operation.
115
Reversal of heparin activity
Protamine chloride was administered in 19 cases (8 receiving UFH and 11 receiving
LMWH). The dosages varied between 20 and 40 mg. Two patients received protamine
when heparin activity had already disappeared. No heparin was detectable after
protamine administration in UFH-patients. Six of the 12 UFH-patients who did not
receive protamine still had measurable heparin levels (range: 0.1-0.661U/ml) at the end
of operation. Heparin activity was measurable in 9 of the 11 LMWH-patients who had
received protamine (range: 0.05-0.42 IU/ml). At the end of operation heparin was
detectable in all LMWH-patients who had not received protamine (range: 0.05-0.46
U/ml).
Intraoperatively. no thrombotic or bleeding complication occurred in any of the patients.
Postoperatively, two patients of the LMWH group died because of a poor lungfunction
and asystole respectively. One patient in the UFH group underwent relaparotomy
because of postoperative blood loss. The cause of blood loss was not found. The anti
Xa activity at the end of operation in this patient was 0.65 U/ml and the corresponding
APTT value was more than 300 seconds. The preoperative APTT value in this patient
was within the normal range.
The APTT value intraoperatively before heparin administration was 47.3 seconds. Five
minutes after heparinization the APTT value reached the maximum detectable value of
300 seconds and remained on this level until the end of operation.
DISCUSSION
Bypass grafting has become the most widely applied and effective method of treating
aonoiliac disease25• In aonobifemoral bypass grafting systemic heparinization is
common practice that attempts to prevent distal thrombotic complications during and
after temporary occlusion of the aorta. The use of heparin in these procedures is mainly
based on empirical grounds and its value has never been evaluated in a controlled
prospective uial.
Intraoperative heparin monitoring showed that the current methods for administering
heparin may not provide an adequate degree of anticoagulation in all patients24"7
•
116
Information gained from heparin monitoring provides insight into the actual level of
anticoagulation and may be used to predict the amount of additional dosages of heparin
if required. Whether this is effective in reducing the risks of complications due to large
individual differences in heparin response and elimination rate needs to be determined.
Another approach to overcome some of the heparin related complications may be the
use of heparins with different pharmacological properties. In the current study we
compared the pharmacokinetics of intravenously administered LMWH and UFH during
aortobifemoral bypass prosthesis operations. At five minutes after heparin
administration, no difference was observed in the anti-Xa activities induced by UFH
and LMWH. The interindividual variation of anti-Xa activity in the LMWH group was
smaller (SD:0.23) than in the UFH group (SD:0.43) and the anticoagulant effect,
expressed as anti-factor Xa activity, was more sustained in the LMWH group: at 110
minutes after administration, the levels were still in the therapeutical range of 0.2-0.5
IU/ml as defined by others26.27.
Surprisingly. the disappearance curves of the two heparins used in this study showed
quite a similar shape as measured with the anti-Xa assay. This has not been reported
before.
We are aware of only one study that compared the effect of a LMWH and UFH during
aortic graft surger/8. In that study heparin was not administered intravenously before
clamping, but injected into the aorta just distal to the clamp, which resulted in delayed
appearance of peak levels of heparin activity.
Which conditions existing in this study could have affected the pharmacokinetic
behaviour of the heparin fragments? Firstly. shortly before completion of the
anastomosis, the graft was flushed with variable amounts of a fluid containing 25 IU
of UFH per ml normal saline solution. Since the two heparin assays employ different
calibration curves, it is possible that the flushing of the graft is only detected in the
UFH calibration curve (see arrow in figure I) and hence may influence the
pharmacokinetic profile of UFH rather than that of LMWH.
Secondly, the pharmacokinetics of the two heparins probably have a different metabolic
basis. It has been suggested that elimination of heparin from the circulation is the result
ll7
of a combination of a saturable and a linear clearance curve". The saturable mechanism
involves binding of heparin to endothelial cells and uptake into the reticulo-endothelial
system"·". The non-saturable mechanism (""linear clearance") could be related to renal
clearance32• High molecular weight forms of heparin are preferably eliminated by the
saturable mechanism". while LMWH is predominantly cleared from the circulation by
the non-saturable renal route34• This model is supponed by the observation in patients
that severe chronical renal insufficiency did not alter the half-life of UFH, but strongly
prolonged that of LMWH35• Also. the prolongation of the half-life of LMWH in
nephrectomized rabbits was much more pronounced than that of UFH36• Therefore it
is conceivable that the large amounts of fluids infused during aonic surgery could have
affected the renal clearance route of heparin. which would especially alter the
pharmacokinetic behaviour of LMWH and not that of UFH. This is in agreement with
the observation by others that the decay curves of UFH in patients undergoing vascular
surgery are similar pre-and intraoperatively10•
The finding that LMWH prolongs the APTT to a lesser extent than UFH and that the
APTT values were not correlated with the anti-factor Xa levels in the LMWH group
is in accordance with other stud1es2 1.37 ·3 ~.
FbDP levels in plasma. i.e. cross-linked and non cross-linked derivatives of fibrin.
increased significantly in both groups. This has also been reponed by others using a
different assay specific for cross-linked fibrin degradation products"""'. The
peroperative rise of FbDP levels could be explained by the surgical trauma which is
known to trigger blood coagulation mechanism" but may also be caused by lysis of e.g.
intravascular thrombi. The large standard deviation of the FbDP levels in the LMWH
group (6.01 ug/ml) can be explained by the observation of very high FbDP values in
one patient (13.2 ug/ml before heparin administration and 28.0 ug/ml after removal of
the clamp). No clinical abnormalities were found in this patient.
This study confirms earlier repons that administration of protamine on empirical basis
is inadequate4•11 and that the anti-factor Xa activity of LMWH is. in contrast to that of
UFH. not completely reversible by protamine (detectable levels ranged from 0.05 to
0.42 IU/ml). This has clinical relevance because observations from clinical studies
H8
showed that plasma concentrations exceeding 0.4 anti-factor Xa !U/ml produce a
critically enhanced risk ofbleeding44.ln our study no bleeding accident occurred in any
of the patients receiving LMWH and the blood loss was comparable in both groups.
In conclusion, following an intravenous bolus injection, the pharmacokineric profiles
of LMWH and UFH show a similarity in patients undergoing aonobifemoral bypass
grafting.
REFERENCES
I Murray G. Heparin in surgical treatment of blood vessels. Arch Surg
1940.40:307-325
2 Jacobsen W. Brauer F. Smith LL. Heparin activity monitoring during vascular
surgery. Am J Surg 1978:136:141-144
3 Quigley FG. Jamieson GO, Lloyd JV, Faris lB. Monitoring of heparin in
vascular surgery. J Vase Surg 1988:8:125-127.
4 Pone RJ. De Jong E, Knot EAR. De Maat MPM. Terpstra OT. Van Urk H,
Greenland THN. Monitoring heparin and haemostasis during reconstruction of
the abdominal aona. Eur J Vase Surg 1987;1:397-402
5 Robinson MHE, Studley JON. Powis SJA. Anticoagulation in abdominal aortic
aneurysm surgery: the approach of vascular surgeons in Great Britain and
Ireland. Eur J Vase Surg 1989;3: 141-143
6 Collins OJ. Kimbal DB. Rich NM. Andersen CA. McDonald PT. Heparin
utilization during anerial vascularization. Am J Surg 1978:44:753-757
7 Manny J, Romanoff H. Hyamm E. Manny M. Monitoring of intraoperative
heparinization in vascular surgery. Surgery 1976;89:641-643
8 Barner HB. The use of heparin in anerial reconstructions. Surg Gynecol Obstet
1974;138:920
9 House AK. Potter JM. Smith PA, Kennet D, Chester A. Heparinization in aortic
surgery. J Cardiovasc Surg 1988:29:707-711.
10 Williams NN, Brae PJ. Burke P. Meagher EA. o· Donoghue C, Otridge B,
Bouchier-Hayes D. Heparin kinetics in vascular surgery. Eur J Vase Surg
119
1989;3:493-469.
II Mabry CD, Thompson BW, Read RC. Activated clolling time (ACT) monitoring
of intraoperative heparinization in peripherical vascular surgery. Am J Surg
1979;138: 894-900
12 Heimbecker RO. Systemic heparin in major vascular operations (editorial). Surg
Gynecol Obstet. 1977;144:735.
13 Effeney DJ, Goldstone J, Chin D. Krupski WC, Ellis RJ. Intraoperative
anticoagulation in cardiovascular surgery. Surgery 1981;90(6):1068-1074
14 De Swart CAM, Nijmeijer B, Roelofs JMM, Sixma JJ. Kinetics of intravenously
administered heparin in normal humans. Blood 1982;60:1251
15 Lane DA, Denton J, Flynn AM, Thunberg L, Lindahl U. Anticoagulant activities
of heparin oligosaccharides and their neutralization by platelet factor 4. Biochem
J 1984;218:725
16 Bratt G. Tomebohm E. Widlund L. Lockner D. Low molecular weight heparin
(Kabi 2165. Fragmin): pharmacokinetics after intravenous and subcutaneous
administration in human volunteers. Thromb Res 1986; 42: 613-620.
17 Bakker M. Dekker PJ, Knot EAR, Van Bergen PFMM, Jonker JJC. Home
treatment for deep venous thrombosis with low molecular weight heparin.
Lancet 1988, nov 12:1142 (letter).
18 Levine MN, Hirsch J. An overview of clinical trials with low molecular weight
heparin fractions. Acta Chir Scand 1988;154 (suppl 543):73-79
19 Harenberg J. Gnasso A. de Vries J, Zimmermann R. Augustin I. Anticoagulant
and lipolytic effects of a low molecular weight heparin fraction. Thromb Res
1985 ;39: 683
20 Bratt G, Tomebohm E. Lockner D. Bergstrom K. A human pharmacological
study comparing conventional heparin and low molecular weight heparin
fragment. Thromb Haemost 1985;53:208
21 Lockner D, Bratt G, Tornebohm E, Aberg W, Granqvist S. Intravenous and
subcutaneous administration ofFragmin in deep venous thrombosis. Haemostasis
1986;16 (suppl 2):25-29.
120
22 Teien AN, Lie M, Abildgaard U. Assay of heparin in plasma using a
chromogenic substrate for activated factor X. Thromb Res 1976;8: 413-420
23 Bascu D. Gallus A, Hirsch J, Cade J. A prospective study of the value of
monitoring heparin treatment with the activated thromboplastin time. N Eng J
Med 1972;287:324-327
24 Kopperl PW. Hoegee-de Nobel E, Nieuwenhuizen W. A monoclonal antibody
based enzyme immunoassay for fibrin degradation products in plasma. Thromb
Haemostas 1988;59:310-315.
25 Brewster DC, Darlit:g RC: Optimal methods of aortoiliac reconstruction. Surgery
1978: 84: 739 - 748.
26 Hirsch J, van Aken WG, Galius AS et al. Heparin kinetics and pulmonary
embolism Circulation 1976:53:691-695
27 Penner JA. Experience with a thrombotic clotting time assay for measuring
heparin activity. Am J Clin Pathol 1974;61 :645-653.
28 Melissari E, Stringer MD, Kakkar VV. The effect of a bolus injection of
unfractionated or low molecular weight heparin during aortobifemoral bypass
grafting. Eur J Vase Surg 1989;3:121-126
29 Glimelius B, Busch C, Hoo M. Binding of heparin on the surface of cultured
human endothelial cells. Thromb Res 1978:12:773
30 Saba Hl, Saba SR. Morelli GA. Anti-heparin activity in human cells. Thromb
Res 1986:42:355
31 Mahadoo J, Hiebert J. Jacques LB. Vascular sequestration of heparin. Thromb
Res 1977;12:79
32 Piper J. The fate of heparin in rabbits after intravenous injection. Filtration and
tubular secretion in the kidney. Acta pharmacologica 1947;3:373
33 Palm M, Mattson Ch. Pharmacokinetics of heparin and low molecular weight
heparin fragments (Fragmin) in rabbits with impaired renal and hepatic function.
Thromb Haemostas 11987;58:932
34 Boneu B. Caranobt. C. Gabaig AM, Dupouy D. Sie P. Buchanan MR, Hirsch
J. Evidence for a saturable and reversible mechanism of clearance of standard
121
heparin in rabbits. Thromb Res 1987;46:845
35 Goudable C, Ton That H, Damani A, Durand D, Caranobe C, Sie P, Boneu B.
Low molecular weight heparin half life is prolonged in haemodialysed patients.
Thromb Res 1986;43: I
36 Caranobe C, Barret A, Gabaig AM, Dupouy D, Sie P, Boneu B. Disappearance
of circulating anti-Xa activity after intravenous injection of standard heparin and
low molecular heparin (CY 216) in normal and nephrectomized rabbits. Thromb
Res 1985;40: 129
37 Walenga JM, Fareed J, Petitou M, Samama M, Lormeau, JC, Choay J.
Intravenous antithrombotic activity of a synthetic heparin penta saccharide in a
human serum induced stasis thrombosis model. Thromb Res 1986; 43: 243
38 Caner CJ. Kelton JG, Hirsch J, Cerkus A, Santos AV, Gent M. The relationship
between the hemorrhagic and antithrombotic properties of low molecular weight
heparin in rabbits. Blood 1982;59(6):1239-1245
39 De Prost D, Ollivier V, Vie P, Benacerraf R, Duparc J, Khoury A. D-dimer and
thrombin-antithrombin II complex levels uncorrelated with phlebographic
findings in I 1 total knee replacement patients. Ann Bioi Clin 1990:48:235-238
40 Jorgensen LN, Lind B. Hauch 0, Leffers A, Albrecht-Beste E, Konradsen LAG.
Thrombin-antithrombin Il-complex and fibrin degradation products m
plasma:surgery and post operative thrombosis. Thromb Res 1990;59:69-76
41 Hauch 0, Jorgensen LN, Kolle TR. Wille-Jorgesen P, Nerstrom H, Neerstrand
HS. Plasma cross-linked fibrin degradation products fraction D in patients
undergoing elective abdominal surgery. Thromb Res 1988;51 :385-389
42 Johnson EJ, Hariman H, Hampton KK, Grant PJ, Davies JA, Prentice CRM.
Fibrinolysis during major abdominal surgery. Fibrinolysis 1990;4:147-151
43 Grant PJ, Tate GM, Davies JA, Williams NS, Prentice CRM. Intraoperative
activation of coagulation-a stimulus to thrombosis mediated by vasopressin?
Thromb Heamost 1986:55:104-107
44 Hirsch J, Ofosu FA. Levine M. The development of low molecular heparins for
clinical use. In: Verstraete M. Yermylen J, Lijnen HR, Arnout J, eds.
122
Thrombosis and Haemostasis: Leuven University Press, 1987: 325-348.
123
Chapter 6
DIAGNOSTIC VALUE OF QUANTITATIVE TESTS FOR FIBRIN
DEGRADATION PRODUCTS IN DEEP VENOUS THROMBOSIS
H. Kroneman'~. M. Berger', P.M.M. Bossuyt1, E.A.R. Knot2
,
J.H.P. Wilson', J.D.F. Habbema'
Center for Clinical Decision Analysis', Erasmus University Rotterdam, Department of
Internal Medicine If, University Hospital Rotterdam-Dijkzigt, Rotterdam, The
Netherlands
submitted for publication
SUMMARY
Laboratory diagnosis of deep venous thrombosis (DVT) has regained interest through
development of enzyme immuno assays (EIA's) based on monoclonal antibodies for
detection of fibrin degradation products (FbDP's) in plasma, This study was undertaken
to evaluate the diagnostic value of these new assays in the diagnosis of DVT on the
basis of reponed data in literature using receiver operating characteristics (ROC)
methodology and Bayesiar1 analysis, Comparing test characteristics of the D-dimer
assay for DVT at cutoff values as reponed in literature showed a significant difference
between test performance in hospitalized and outpatients, By applying Bayes' rule on
pooled data of hospitalized and outpatients separately, the relationship was calculated
between D-dimer values, pre-test and post-test probabilities of DVT, It was concluded
that test performance of the D-dimer assay for DVT is different in outpatients and
hospitalized patients. Current diagnostic concepts concerning FbDP assessment in
patients suspected for deep venous thrombosis should be adapted,
126
INTRODUCTION
Laboratory assessment of fibrin and fibrinogen degradation products (FDP's) in deep
venous thrombosis (DVT) has been practised for several years using different
techniques (1-7). Most of these assays could only be performed on serum samples using
polyclonal antibodies which cross-react with fibrinogen. Because serum is a source of
artefacts (8), results obtained by these serum-based assays for FDP's should be
interpreted with caution (9). As a consequence, the initial suggestion that determination
of FDP's in serum could be of value in the diagnosis of DVT was abandoned.
The issue of measuring FDP's in patients with DVT has regained interest through the
development of quantitative enzyme immuno assays (EIA's) for their assessment (10-
15). These EIA ·s employ monoclonal antibodies against specific fibrin and fibrinogen
epitopes which do not crossreact with fibrinogen. Furthermore, these assays permit
discrimination between degradation products of fibrin (FbDP's) and fibrinogen
(FgDP's) and can be performed on plasma samples, thereby avoiding the problems
inherent in the use of serum. Clinical experience with the new assays is steadily
growing and it is becoming clear that diagnostic concepts concerning FbDP assessment
in patients suspected for DVT may have to be adapted. Therefore an evaluation of the
diagnostic characteristics of these assays in patients suspected for DVT seems
appropriate.
Diagnostic tests are often characterized using the indices of sensitivity, which is the
probability that the test result is positive in patients who have the disease (also called
true positive rate) and spe~ificity, which is equivalent to the probability that the test is
negative in patients who do not have the disease (also called true negative rate). Most
of the published studies concerning DVT and the new assays for FbDP assessment have
reported sensitivity and specificity based on single cutoff points to distinguish between
diseased and non-diseased. The use of different cutoff points makes direct comparisons
between studies difficult. Furthermore, the appropriate cutoff value in clinical practise
depends on the patient population in which the test is used.
By dichotomization of test results, the test is either positive or negative. valuable
information may be lost, eg. suppose a hypothetical cutoff point of 20; in this case a
127
test result of 10.000 will be interpreted in exactly same way as a value of 21, although
the probability of diseas~ may be considerably higher. A useful parameter to
characterize clinical information without dichotomization is the likelihood ratio (LR),
which is the ratio of the probability of a test result in diseased persons and the
probability of the same test result in nondiseased persons (21). Another way, although
also based on dichotomization, to obtain more information upon a diagnostic test is to
determine sensitivity and specificity for different cutoff points. This is possible with
EIA' s for FbDP' s since titey are quantitative tests yielding results distributed on a
continuous scale, which may be reduced to an ordinal scale by categorizing the test
variable. In this way, a variety of pairs is generated which can be plotted as a set of
points in a unit square. This constitutes a receiver operating characteristic (ROC) curve.
The area under a ROC cu·-ve is a measure of test performance and may be used for
comparing tests (16-20).
The post-test probability of disease can dramatically be affected by the prevalence or
pre-test probability of disease: increase or decrease of disease prevalence will be
accompanied respectively by a rise and fall of post-test probability. Hence, when
interpreting test results oue should be aware of disease prevalence in the studied
population. The relation between test result, pre-test and post-test probability of disease
can be determined with Bayes' Theorem (21,22).
The aim of this study was to evaluate the diagnostic value of the new assays for DVT
by use of ROC methodology and Bayesian analysis on published data obtained in
patients presenting with symptoms suggesting DVT.
MATERIALS AND METHODS
Materials
The English language medical literature was searched for articles that have assessed
diagnostic test characteristics ofElA's for fibrin degradation products in DVT. Because
the first report that described an E!A for fibrin degradation products based on
monoclonal antibodies was published in 1983 (II), we conducted a MEDUNE search
128
from january 1983 to january 1991 for articles discussing quantitative assessment of
fibrin degradation products in plasma and deep venous thrombosis. In addition we
reviewed the bibliographies of the articles found by the above method. Each article
resulting from this search was screened and assessed for study design. Our criteria for
inclusion of the articles in this study were: a blind comparison of the results of the
index test with the current gold standard (ascending venography); a prospective
evaluation of consecutive patients presenting with signs or symptoms suggesting DVT;
a blind interpretation of test results and clear definition of criteria for test positivity of
both index test and gold standard test.
Methods
ROC curves were constructed on the basis of data presented in each study fulfilling the
inclusion criteria. True positive rates (or sensitivity) were plotted on the abcis~ false
positive rates (or I minus specificity) on the ordinate. Cutoff points were chosen in
such a way that the pairs of true-positive rates and false negative rates were evenly
spaced along the ROC curves. The area under the ROC-curve was calculated by
connecting the calculated points and adding the areas of the trapezoids underneath each
pan of the curve. The standard errors associated with the areas were determined
according to the Hanley-M:Neil algorithm (17). In comparing the area under two ROC
curves, the two sample t-test was used. Within studies that determined test
characteristics for hospitalized and outpatients separately. a comparison was made
between the proportions of nonDVT inpatients and nonDVT outpatients with D-dimer
levels exceeding the reported cutoff value. In testing differences between proportions,
the Chi-square test for independent samples was used. All reported significance levels
were two-sided and any probability less than 0.05 was considered to represent a
significant difference.
Post-test probabilities of DVT were calculated for different prevalences of DVT by use
of the Odds form of Bayes' rule (21.22) which can be written as:
P(DVT[T) P(noDVT[T)
P(OVT) x P(T[ DVT) P(noDVT) P(T[noDVT)
129
where P(DVTIT) is the post-test probability of DVT given a test result T and P(DVT)
is the prevalence of DVT in the population under investigation; P(TIDVT) is the
probability of a test result given the presence of DVT; P(noDVTIT) is the probability
of DVT absence given a test result and P(TinoDVT) is the probability of finding a test
result given the absence of DVT. The quotient of P(TIDVT) and P(TinoDVT) is called
the likelihood ratio (LR).
RESULTS
From a total of fifteen retrieved papers (13,14,23-35). four studies met our criteria for
inclusion (32-35). Nine papers were not included in our study because test performance
was evaluated in a population consisting of patients with confirmed DVT (!3,14,23-29),
one because impedance plethysmography was used to confirm DVT (30) and one
because only results of the semiquantitative latex agglutination assay were reponed
(31). The results of the published reports that were included in our study are
summarized in Table I. All studies used the 0-dimer assay for assessment of FbDP
plasma levels. Results were calculated for hospitalized and outpatients separately in
case this had not been done by the authors and could be extracted from the available
data. Figure l shows the relative frequency distributions of the categoric D-dimer
values in outpatients and hospitalized patients separately. As can be seen, most
outpatients without OVT has lower D-dimer levels compared with nonOVT inpatients,
whereas 0-dimer levels of inpatients with confirmed DVT were in general higher than
those of outpatients. The areas of the ROC curves of individual studies obtained in
outpatients were statistically not different (0.2 <p< 0.35). The same was true for the
ROC areas of inpatients (p=0.23). Data of individual studies were pooled for
hospitalized patients and outpatients separately. Comparison of the areas of the two
pooled ROC curves showed that the ROC curve for inpatients (area: 0.80 ± 0.05) was
130
Table 1 Test characteristics of the D-dimer assay in patients suspected for DVT as reported in
literature
Size Population (ref) Capture Cutoff value Prevalence Sensitivity Specificity
antibody (ng/ml) of DVT
50 inpatients (32) DD-386/22 250 52% 100% 17%
54 outpatients (32) DD-386/22 250 35% 100% 66% -108 outpatients (33) DD-386/22 200 36% 97% 65% "' -108 outpatients (33) anti-FDP-D.D. 500 36% 97% 62%
25 outpatients (34) anti-FDP-D.D. 200 20% 100% 50%
25 outpatients (34) anti-FDP-D.D. 500 20% 86% 72%
28 inpatients (34) anti-FDP-D.D. 200 50% 100% 7%
28 inpatients (34) anti-FDP-D.D. 500 50% 100% 14%
58 in+outpatients (35) DD-386/22 200 41% 100% 47%
Figure 1
Proportion 0.7
0-200
Proportion 0.7
0.6
0.5
0.4
0.3
0.2
0.1
Outpatients g Normals
Ei! (noo122.)
D ovr (n • 65)
200-500 500-1000 1 000-2000 > 2000
0-dirner (ng/rnl)
Inpatients g Normals
e (n .. 38)
Dovr (n = 40)
0 '--""'o"'-2'-o-o-- 2oo-5oo 5oo-1 ooo 1 ooo-2ooo > 2""o-oo-'--O-dirner (ng/rnl)
Relative frequencies of categoric D-dimer values in outpatients and inpatients
with symptoms suggesting DYT.
not as good as for outpatients (area: 0.86 ± 0.03). although the difference was
statistically not significant (p=0.14). The area of the ROC curve based on reported data
obtained in a mixed population of hospitalized and outpatients (35) was not
132
significantly different as compared with that of pooled ROCs of inpatients (p=0.48) and
outpatients (p=0.16). The pooled ROC curves are shown in Fig 2.
Figure 2
Sensitivity (true positive rate)
0.9
500
//,// ,///
, .. /~ooo ... ···········/
200
0.8
0.7
.,..-"""" ./ uninformative test 0.6 / .·
:: r:- ///// :: ',///
Outpatients -e-
Inpatients ·-*--
o·~'--'-~--'--'-J.._'--'-~--'--'-J.._'--'-~--'--'-J.._'--'
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
1 - Specificity (false positive rate)
Roc curves of D-dimer in outpatients and inpatients for detection of DVT.
Two studies determined test characteristics of the D-dimer assay for hospitalized and
outpatients separately (32.34 ). One of these studies showed that the proportion of
nonDVT patients with D-dimer levels above the reported cutoff value of 250 ng/ml was
significantly higher (X2= 11.48. p<O.OO I) in the subgroup of inpatients when compared
with outpatients (32). This finding was confirmed in the other study (34) using cutoff
values of 200 ng!ml (X2=4.89, p<0.05) and 500 ng/ml (X2=6.64, p<O.OI). The
relationship between D-dimer value. pre-test and post-test probabilities of DVT is
shown in Table 2. Because of the relatively small sample size of inpatients, LR's in this
133
A. Outpatients
Pre-test Post-test probability of DVT
probability D-dimer levels
ofDVT <200 200-500 500-1000 1000-2000 >2000
10% 0.3% 14% 21% 31% 56% 15% 0.4% 20% 30% 42% 67% 20% 0.6% 27% 38% 50% 74% 25% 0.8% 33% 45% 57% 79% 30% 1% 38% 51% 63% 83% 40% 2% 49% 62% 73% 89% 50% 3% 59% 71% 80% 92% 60% 4% 69% 79% 86% 95% 70% 6% 77% 85% 90% 96%
B. Inpatients
Pre-test Post-test probability of DVT
probability D-dimer levels
ofDVT <200 200-500 500-1000 1000-2000 >2000
10% 1% 2% 5% 11% 86% 15% 2% 4% 7% 16% 9!% 20% 3% 5% 10% 22% 93% 25% 3% 7% 13% 27% 95% 30% 4% 9% 16% 32% 96% 40% 7% 13% 23% 43% 97% 50% 10% 18% 3!% 53% 98% 60% !4% 25% 40% 62% 99% 70% 20% 34% 51% 72% 99%
Table 2 Relationship between D-dimer levels (ng/ml), pre-test and post-test probability
of DVT in o-utpatients (A) and inpatients (B) presenting with symptoms
suggesting DVT.
subgroup were obtained after applying logistic regression analysis on the square roots
of the values. As can be seen from table 2, a rise of disease prevalence is accompanied
by an increase of disease probability in each categoric D-dimer value. Funhermore, the
presence of disease becomes more probable with higher D-dimer values for each of the
134
pre-test probabilities.
DISCUSSION
This study was undertaken to determine the role of measuring FbDP's plasma levels
in the diagnosis of DVT. We used ROC methodology and Bayesian analysis on
published data to assess diagnostic test performance of EIA's for FbDP's in patients
clinically suspected of having DVT. ROC analysis is a convenient method for
comparing diagnostic test performance and may provide insight into the practical use
of diagnostic tests, because it shows various pairs of sensitivity and specificity which
can be obtained by applying different cutoff values. Although test characteristics are
claimed to be independent of pre-test probability, reponed sensitivities and specificities
differ greatly between inpatients and outpatients (Table I). After transforming data into
ROC curves, it was demonstrated that the areas of individual studies obtained in
outpatients and inpatients were comparable, which suggests that test results are
comparable within these subgroups.
As mentioned. the area of the pooled inpatients ROC curve was smaller as compared
with that of outpatients. although this difference was not significant. Comparing
sensitivities and specificities in hospitalized and outpatients at reported single cutoff
points showed, however, that test performance of the D-dimer assay was not
comparable between these subgroups. This is illustrated in Table 2: In case the
prevalence of DVT is 20%, the finding of a D-dimer value between 200 and 500 ng/ml
predicts the presence of DVT in outpatients with a probability of 27%, whereas the
probability for DVT in hospitalized patients is only 5%. The D-dimer distributions of
both DVT and nonDVT inpatients are shifted to the right when compared to outpatients
(Fig. I). A possible explanation for this difference may be a higher incidence of co
morbodity interfering with the coagulation system in hospitalized patients suspected for
DVT.
In the diagnostic process the clinician, after having been supplied with a numerical
value of the D-dimer test result, is interested in the probability of DVT. This post-test
probability, however. depends on two conditions: I. given a certain test value, the post-
135
test probability of disease will change in the same direction with the prevalence. 2.
given a certain pre-test probability. the probability of disease varies with the test value.
The relation between test result, pre- and post-test probability can be calculated with
Bayes' rule and is illustrated in Table 2. Suppose an outpatient with a pre-test
probability for DVT of 20% (prevalence in outpatients); in this case the finding of a
low D-dimer value, eg <200 ng/ml, implies that the probability of DVT reduces to
0.6% (i.e. the probability for DVT absence is± 99%), whereas a D-dimer value of
1500 ng/ml indicates the absence of DVT with a probability of 50%. Even so, in case
of a fixed D-dimer result of 400 ng/ml, the probabilities for absence of DVT may range
from 23% to 86%. depending on the pre-test probability (10%-70%) in the studied
outpatient population (Table 2). After the test result is known. the physician must
choose between doing nothing (ruling out the diagnosis). obtain additional diagnostic
information or treat. The decision to withhold therapy depends on the no-treatment
threshold. which is a measure of the balance between the benefits of treatment to
diseased patients (B) and the risks of treatment to healthy patients (R). The optimal
cutoff point of a diagnostic test can be determined by applying Bayes' rule and the
specification of a risk/benefit ratio. The latter is difficult to estimate because life
expectancy and quality of life should also be taken into account in assessment of a
treatment risk/benefit ratio. However. it is obvious that the treatment threshold
probability for DVT is low. i.e. it has high benefit to patients with DVT (prevention
of fatal pulmonary embolism) and relatively low risk (bleeding) to nonDVT patients.
The relationship between post-test probability for DVT and the risk/benefit ratio is
described in the following formula:
P(DVTIT) = R P(noDVTIT) B
Suppose the benefit of treatment is 5 times higher than the risk, then the threshold post-
136
test probability for DVT is ± 17%. The corresponding cutoff value can now be derived
from Table 2 for each pre-test probability of DVT, e.g. in hospitalized patients with a
pre-test probability of 40%, the appropriate cutoff value would be 500 ng!ml; the
treatment threshold is not reached in case the pre-test probability for DVT is 70%. The
relationship between post-test probability and D-dimer values for the above mentioned
pre-test probabilities of DVT in hospitalized patients is shown in Figure 3.
Post-test probability of DVT 1 ,/:::::::f"c"''_,,
0.8
Pre-test probability of DVT .? __ .... ·····~/ .4
0.6
0.4
0.2
__________ //:/// ,''
---
o~L-----J---~~~~~~L-----J---~~~--~~ 500 1.000 2.000 5.000 10,000 100
Figure 3
200
D-dimer (ng/ml)
Relationship between post-test probability and D-dimer levels (ng/ml) at fixed
pre-test probabilities of DVT in hospitalized patients.
This is just an example of how a cutoff point could be assessed and does not
necessarily imply that this should be the optimal cutoff point for clinical use. The
appropriate cutoff point depends on the patient population in which the test is used: eg.
the pre-test probability of DVT in hospitalized patients after total hip replacement is
considerably higher as compared with that of outpatients, hence the cutoff point
137
necessary to obtain an equal post-test probability for absence of DVT must be higher
in the hospitalized group.
In conclusion, evaluating D-dimer test results with ROC methodology on reported data
demonstrated that D-dimer test characteristics were comparable for individual studies
in outpatients and hospitalized patients separately. The D-dimer assay performs
significantly different in outpatients when compared with hospitalized patients in the
diagnosis of DVT at reported cutoff points.
By applying Bayes theorem on pooled data of in- and outpatients, posterior
probabilities for DVT could 3be calculated in relation to categorized D-dimer values
and different prior probabilities. Based on this analysis, the optimal cutoff point can be
calculated by combining Bayes rule and estimating a treatment risk/benefit ratio. A
reconsideration of the clir.ical utility of FbDP assessment in the diagnosis of DVT
seems warranted on the basis of available evidence on test characteristics.
REFERENCES
I. Ruckley CV. Das PC, Leitch AG.Donaldson, AA. Copland WA, Redpath AT,
Scott P. Cash JD: Serum fibrin/fibrinogen degradation products associated with
postoperative pulmonary embolus and venous thrombosis. Brit Med I 1970;
4:395-3
2. Redner U, Nilsson I M: Clinical experience with determination of fibrinogen
degradation products. Acta Med Scand 1971; 189:471-477.
3. Wood EH. Prentice CRM. McNicol GP: Association of fibrinogen - fibrin
related antigen (F.R. antigen) with postoperative deep vein thrombosis and
systemic complications. Lancet 1971; i:I66-I69.
4. Gallus AS. Hirsch J, Gent M: Relevance of preoperative and postoperative blood
tests to postoperative leg vein Thrombosis. Lancet 1973; ii:805-809.
5. Gurewich V, Hume M, Patrick M: The laboratory diagnosis of venous
thromboembolic disease by measurement of fibrinogen/fibrin degradation
products and fibrin monomer. Chest 1973; 64;585-590.
6. Cooke ED, Gordon YB. Bowcock SA. Sola CM, Pilcher MF, Chard T, Ibbotson
138
RM, Ainsworth ME: Serum fibrin(ogen) degradation products in diagnosis of
deep vein thrombosis and pulmonary embolism after hip surgery. Lancet 1975:
ii:51-54.
7. Clayton JK, Anderson JA, McNicol GP: Preoperative prediction of postoperative
deep vein thrombosis. Brit Med J 1976; 2:910-912.
8. Nieuwenhuizen W. Plasma assays for derivatives of fibrin and fibrinogen, based
on monoclonal antibodies. Fibrinolysis 1988;2: 1-5
9. Gaffney PJ, Perry MJ: Unreliability of current serum fibrin degradation
products(FDP) assays. Thromb Haemost 1985; 53:301-302.
10. Koppert PW. Kuipers W, Hoegee-de Nobel E, Brommer EJP, Koopman J,
Nieuwenhuizen W: A quantitative enzyme immunoassay for primary
fibrinogenolysis products in plasma. Thromb Haemostas 1987; 57:25-28.
ll. Elms MJ. Bunce JH. Bundesen PG. Rylatt DB, Webber AJ, Masci PP, Whitaker
AN: Measurement of crosslinked fibrin degradation products- An immunoassay
using monoclonal antibodies. Thromb Haemostas 1983; 50:591-594.
12. Rylan DB. Blake AS, Glottis LE, Massingham DA, Fletcher WA, Masci PP,
Whitaker AM. Elms M. Bunce J. Webber A. Wyatt D. Bundesen PG: An
immunoassay for human D-dimer using monoclonal antibodies. Thromb Res
1983; 31:767-778.
13. Koppert PW. Hoegee-de Nobel E, Nieuwenhuizen WA: monoclonal antibody
based enzyme immunoassay for fibrin degradation products in plasma. Thromb
Haemostas 1988; 59:310-315.
14. Elms MJ. Bunce lA, Bundesen BG. Rylatt DB, Webber AJ. Masci PP, Whitaker
AN: Rapid detection of cross-linked fibrin degradation products in plasma using
monoclonal antibody-coated particles. Am J Clin Path 1986: 85:360-364.
15. Koopman J. Haverkate F. Koppert PW. Nieuwenhuizen W, Brommer EJP, van
der Werf WGC: New enzyme immunoassay of fibrin-fibrinogen degradation
products in plasma using a monoclonal antibody. J Lab Clin Med 1987; 109:75-
84.
16. Metz CE. ROC methodology in radiological imaging. Invest Radio!
139
1986;21 :720-733
17. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver
operating characteristic curve. Radiology 1982; 143:29-36
18. Hanley JA, McNeil BJ. A method of comparing the areas under receiver
operating characteristic curves derived from the same cases. Radiology
1983;148:839-843
19. McNeil BJ, Hanley JA. Statistical approaches to the analysis of receiver
operating characteristic (ROC) curves. Med Decis Making 1984;4: 137-150
20. McClish DK. Comparing the areas under more than two independent ROC
curves. Med Decis Making 1987;7:149-155
21. Sox HC, Blatt MA, Higgins MC, Marton Kl. Medical Decision Making. Boston,
Butterworths Publish; 1988: 67-98.
22. Sackett DL. Haynes RB, Tugwell P. Clinical epidemiology. A basic science for
clinical medicine. Little, Brown and Company. Boston!foronto. lsted.:ll8-119
23. Faivre R, Mirshahi M. Ducellier D, Soria C. Soria J, Mirshahi M, Kieffer Y,
Bassand JP. Caen J, Maurat JP. Evolution of plasma specific fibrin degradation
products during thrombolytic therapy m patients with thrombo-embolism.
Thromb Res 1989: 50; 583-589.
24. Hillyard CJ, Blake AS, Wilson K. Rylatt DB. Miles S, Elms MJ, Barnes A,
Bundesen PG. Latex agglutination assay for D-dimer: Evaluation and application
to the diagnosis of thrombotic disease. Clin Chern 1987; 33: 1837-1840.
25. Mirshahi M, Soria C, Soria J, Mirshahi M, Faivre R, Kieffer Y, Bassand JP,
Toulemonde F, Caen J. Changes in plasma fibrin degradation products as a
marker of thrombus evolution in patients with deep venous thrombosis. Thromb
Res 1988: S 1: 295-302.
26. Whitaker AN. Elms MJ, Masci PP, Bundesen PG, Rylatt DB. Webber AJ,
Bunce !H. Measurement of cross linked fibrin derivatives in plasma: an
immunoassay using monoclonal antibodies. J Clio Pathol 1984; 37: 882-887.
27. Speiser W, Mallek R, Koppensteiner R, Stumpflen A. Kapiotis S, Minar E,
Ehringer H, Lechner K. D-dimer and TAT measurement in patients with deep
140
venous thrombosis: Utility in diagnosis and judgement of anticoagulant treatment
effectiveness. Thromb Haemostas 1990; 64(2): 196-201
28. Declerck PJ, Mombaerts P, Holvoet P, De Mol M, Collen D. Fibrinolytic
response and fibrin fragment D-dimer levels in patients with deep vein
thrombosis. Thromb Haemostas 1987; 58(4): 1024-1029.
29. Boisclair MD, Lane DA, Wilde JT, Ireland H, Preston FE, Ofusu FA. A
comparative evaluation of assays for markers of activated coagulation and/or
fibrinolysis: thrombin-antithrombin complex, D-dimer and fibrinogen/fibrin
fragment E antigen. Br J Haematol 1990;74: 471-479.
30. Van Bergen PFMM. Knot EAR. Jonker JJC. De Boer AC. De Maat MPM. Can
quantitative determination of fibrin(ogen) degradation products and thrombin
antithrombin Ill complexes predict deep venous thrombosis in outpatients?
Thromb Haemostas 1989; 62: 1043 - 1045.
31. Wilde JT, KitchenS, Kinsey S. Greaves M, Preston PE. Plasma D-dimer levels
and their relation to serum fibrinogen/fibrin degradation products in
hypercoagulable states. Br J Haematol 1989; 71: 65-70.
32. Rowbotham BJ, Carroll P. Whitaker AN, Bunce lA, Cobcroft RG. Eims MJ,
Masci PP, Bundesen PG, Rylan DB, Webber AJ. Measurement of crosslinked
fibrin derivatives- u~e in the diagnosis of venous thrombosis. Thromb haemostas
1987; 57: 59-61.
33. Ott P, Astrup L. Jensen RH, Nyeland B. Pedersen B. Assessment of D-Dimer
in plasma: Dlagnosdc value in suspected deep venous thrombosis of the leg.
Acta Med Scand 1988; 224: 263-267.
34. Bounameaux H. Schneider PA, Reber G, De Moerloose P. Krahenbuhl B.
Measurement of plasma D-dimer for Diagnosis of deep venous thrombosis. Am
1 Clin Path 1989; 91: 82-85.
35. Heaton DC. Billings JD, Hickton CM. Assessment of D dimer assay for the
diagnosis of deep vein thrombosis. J Lab Clin Med 1987; 110: 588-591.
141
Chapter 7
Discussion
Summary
Samenvatting
Acknowledgements
Curriculum vitae
DISCUSSION
During the last decades insight into the mechanism underlying the formation of fibrin
and its degradation has increased markedly. This progress has provided a basis for the
development of techniques for assessment of the two opposing processes of coagulation
and fibrinolysis.
There are many components in blood which influence the overall fibrinolytic activity
which can be measured in a clinical setting. However, three markers of fibrinolysis
have emerged as being of paramount importance. These are tissue plasminogen
activator (t-PA), plasminogen activator inhibitor (PAl) and fibrin/fibrinogen degradation
products (FDP). While speculation may still exist about the role of measuring activators
and inhibitors of plasminogen, there is agreement that fibrinolysis is reflected by
generation of fibrin and fibrinogen degradation products. Therefore, the most tangible
evidence for ongoing fibrinolysis in blood is the formation of fibrin an fibrinogen
degradation products.
The main difference between the formerly used serum based assays for FDP assessment
and the newly developed &ssays is, that the latter assays can be carried out in plasma
using monoclonal antibodies, which makes it possible to avoid serum artifacts.
Furthermore, these new plasma assays permit discrimination between degradation
products of fibrin and of fibrinogen (chapter 1 ).
We found that fibrinogen degradation products (FgDP) are generally well correlated
with degradation products of fibrin (chapter 2). One may speculate about the origin of
the plasma FgDP fraction: it is unknown whether this FgDP fraction originates from
circulating fibrinogen or the fibrin-fibrinogen complex in the circulation. It is possible
that plasmin. which lysis fibrin. also digests a part of the fibrinogen molecule before
its action is terminated by a,-antiplasmin. The observation of good correlations between
the FgDP fractions and degradation products of fibrin suggests that primary fibrinogen
digestion in the circulation is a frequent event. which is associated with secondary
fibrinolysis.
We examined the role of the D-dimer assay for the diagnosis of deep venous
145
thrombosis (DVT). Analysis of published studies showed that test performance of the
D-dimer assay for DVT at reported cutoff values was different in hospitalized patients
when compared with outpatients (chapter 6). This can be explained by the higher
incidence of co-morbidity interfering with the coagulation system in the first group. It
was demonstrated that very low levels of D-dimer could predict the absence of DVT,
whereas very high levels of D-dimer confirmed the presence of DYT. This finding
suggests that the D-dimer assay may be used for the diagnosis of DVT. but is in
contrast to the results of our study with the D-dimer assay in outpatients, which was
performed at the Thrombosis Service Centre of Rotterdam (chapter 3). The test
characteristics of the D-dimer assay in this study (sensitivity 92%, specificity 21%,
negative predictive value 88% ,positive predictive value of 28%) were disappointing.
This discrepancy may be explained that the fact that lPG was used to confirm DVT
instead of ascending venography, which is less sensitive detecting calf vein thrombosis.
The effectiveness and feasibility of the D-dimer assay for the diagnosis of deep venous
thrombosis can be demonstrated by a management study. Other clinical applications of
the assays for fibrin and fibrinogen degradation products may concern the monitoring
of treatment (chapter 5) and the course of disease (chapter 4 ). In general, detection of
fibrin(ogen) derivatives by monoclonal antibody based assays provides valuable
information on haemostasis in diseases associated with an activated state of coagulation.
146
SUMMARY
This thesis comprises clinical and methodological studies employing newly developed
plasma assays for assessment of fibrin and fibrinogen degradation products (FDP's) on
patients with diseases or conditions characterized by an activated state of coagulation.
In the introduction the mechanism of the haemostatic balance is briefly summarized
based on current insights and it is described that disturbance of this balance may have
two different clinical outcomes: bleeding or thromboembolism. The importance to
obtain information regarding the state of the haemostatic balance in patients at risk for
developing thromboembolism and/or bleeding is emphasized. The aims of the studies
are defined.
In chapter I the disadvantages of existing serum FOP assays. based on polyclonal
antibodies. are described. The newly developed FOP assays employ monoclonal
antibodies which do not cross-react with fibrinogen and can be performed on plasma
samples, thereby avoiding the disadvantages inherent in the use of serum samples. A
summary is given of the currently available FOP assays in plasma and their clinical
relevance has been reviewed as reported in literature. The aims of the studies were
defined on the basis of this review.
In chapter 2 we have evalu~ted four different plasma FOP assays in healthy volunteers,
patients with cirrhosis of the liver and patients suspected for deep venous thrombosis
(OVT). Plasma FOP levels. as assessed by different assays, showed a good correlation
in each (sub)group of patients. In general, the correlations between plasma FOP assays
were lower in normals than in patients. It was first seen that secondary fibrinolysis
appeared to be associated with fibrinogenolysis in cirrhosis of the liver and OVT.
In chapter 3 the diagnostic value of the D-dimer assay for deep venous thrombosis
(OVT) is studied in 232 outpatients clinically suspected of having OVT. The Sensitivity
and specificity of the 0-dimer assay for OVT were 92% and 21% respectively. The
OVT prevalence was 25% and the predictive value of a negative and positive test result
were 88% and 28% respectively. It was concluded that the D-dimer assays has limited
value, either to confirm or to exclude OVT.
147
In chapter 4 the fibrinolytic response in patients with gram-negative septic shock is
studied. In the past, disturbance of haemostasis in septic shock was considered to
depend largely upon extreme activation of the coagulation pathway. It has recently been
suggested that fibrinolysis may be impaired during septic shock. Measurement of
plasma FgDP and TDP plasma levels in patients with septic shock showed that the
fibrinolytic activity was comparable in survivors and nonsurvivors. The process of
ongoing fibrinolysis during septic shock is evidenced by elevated TDP and FgDP
plasma levels. It was concluded from this study that the fibrinolytic system remains
activated during septic shock.
In chapter 5 we focused on heparinization during vascular surgery. During aortic graft
surgery. heparin is administered on empirical basis and is associated with a wide
variation in patient response and elimination rate which makes it difficult maintain
predictable or adequate levels of anticoagulation for all patients. The recently developed
low molecular weight heparins and the FDP plasma assays may be used to place
heparin administration during vascular surgery on a rational scientific basis. We studied
the pharmacokinetics of low molecular weight heparin (LMWH) and unfractionated
heparin (UFH) during aortobifemoral bypass grafting. It was demonstrated that plasma
FbDP levels showed a significant rise intraoperatively in both patients groups, which
could be explained by lysis of intravascular thrombi. FbDP levels correlated moderately
with anti-factor Xa in the LMWH group, but not in the UFH group. Whether this may
be used to define dose regimen for heparin administration needs to be determined by
furtber studies. lt was concluded that the pharmacokinetic profiles of UFH and LMWH
showed a similarity after intravenous bolus injection in patients undergoing
aortobifemoral bypass grafting and that the anti-Xa activity in the LMWH group was
more sustained as compared with the UFH group.
In chapter 6 the value of plasma FDP assays for the diagnosis of disease is evaluated
in a broader sense. This chapter comprises an analysis of published studies on the
diagnostic value of the plasma FDP assays for deep venous thrombosis (DVT). Test
performance of the D-dimer assay for DVT at reported cutoff values was different in
hospitalized patients when compared with outpatients. The relationship between D-
148
dimer levels, pre-test and post-test probabilities of DVT were calculated. Very low
levels of D-dimer could predict the absence of DVT, very high levels of D-dimer
confirmed the presence of DVT. It was concluded that the D-dimer assay may be used
in the diagnosis of DVT and current diagnostic concepts concerning FDP assessment
in patients suspected for DVT should be adapted.
In the discussion has been outlined that assessment of fibrin and fibrinogen degradation
products is the most tangible evidence for ongoing fibrinolysis, and secondary
fibrinolysis appears to be associated with fibrinogenolysis. Whether assays for fibrin
degradation products can be used in daily clinical practise for diagnosis of deep venous
thrombosis can be determined by management studies. Application of the assays on
patients with diseases associated with an activated state of coagulation allows detailed
information on the haemostatic balance.
149
SAMENVATTING
In dit proefschrift worden klinische en methodologische studies beschreven naar fibrine
en fibrinogeen afbraakprodukten in ziekten of condities met een gestoorde haemostase.
De studies zijn uitgevoerd met recent ontwikkelde, op monoclonale antilichamen
gebaseerde enzyme immuno assays.
De inleiding gaat in op bet mechanisme van het haemostase evenwicht. Een verstoring
van deze balans kan twee verschillende klinische gevolgen hebben: bloedingen of
trombo'embolieen. Om therapeutische redenen is bet van belang te weten waar het
evenwicht ligt bij patienten met een verhoogd risico op bloedingen of trombo
embolieen. De doelstellingen van de studie worden uiteengezet.
Hoofdstuk 1 geeft een literatuuroverzicht van de beschikbare laboratoriumtesten voor
fibrine en fibrinogeen afbraakprodukten (FDP). De nadelen van reeds Ianger bestaande
serum FDP assays worden besproken. De recent ontwikkelde FDP testen zijn gebaseerd
op monoclonale antilichamen. die niet kruisreageren met fibrinogeen. en kunnen worden
uitgevoerd met plasma samples. Hierdoor worden de nadelen, welke inherent zijn aan
het gebruik van serum samples. venneden.
In hoofdstuk 2 worden vier verschillende plasma FDP assays geevalueerd in patienten
met levercirrhose, patienten die worden verdacht van diep veneuze trombose (DVT) op
basis van het klinisch beeld, en in niet zieke vrijwilligers. De plasma FDP
concentraties, bepaald door verschillende assays. correleren goed met elkaar in de
verschillende patienten groepen. doch minder goed in de groep van niet zieke
vrijwilligers. Op theoretische gronden mag worden verwacht, dat de berekende som van
de FbDP (fibrine afbraal,produkten) en FgDP (fibrinogeen afbraakprodukten)
concentraties gelijk is aan de bepaalde TDP (totaal aan fibrin and fibrinogeen
afbraal.:produkten) concentratie. Wij hebben echter geen verklaring kunnen vinden voor
de discrepantie tussen de TDP waarden en de berekende som van de FbDP en FgDP
concentraties in 4 patienten met DVT en 1 patient met klinische symptomen van DYT,
waarbij de diagnose niet kon worden gesteld. In deze studie wordt voor het eerst
aangetoond, dat secondaire fibrinolyse gepaard gaat met fibrinogenolyse in levercirrhose
151
en DVT.
In hoofdstuk 3 wordt de diagnostische waarde van de D-dimer assay voor DVT
bestudeerd. De studie werd ge"initieerd omdat er behoefte bestaat aan een simpele niet
invasieve laboratoriumtest voor bet vaststellen van DVT. 232 poliklinische patienten
met verdenking op een trombose been zijn onderzocht. De sensitiviteit en specificiteit
van de D-dimer assay zijn respectievelijk 92% en 21%, de voorspellende waarde van
een positief en negatief test resultaat bedragen respectievelijk 88% en 28% bij een DVT
prevalentie van 25%. Uit het onderzoek wordt geconcludeerd, dat de D-dimer een
geringe diagnostische waarde heeft voor DVT.
In hoofdstuk 4 wordt de fibrinolytische respons beschreven in patiemen met een gram
negatieve sepsis. In het verleden werd de verstoring van de haemostase bij septische
shock voornamelijk toegeschreven aan sterke activatie van de stollingscascade. Recent
onderzoek suggereen. dat de fibrinolyse geremd kan zijn tijdens septische shock. Deze
studie werd uitgevoerd om de mate van fibrinolytische activiteit te bepalen door het
meten van de eindprodukten van de fibrinolyse. Plasma FgDP en TDP concentraties
zijn verhoogd. maar niet significant verschillend in suiVivors en nonsurvivors van een
septische shock. De studie toont aan, dat de fibrinolyse geactiveerd blijft tijdens gram
negatieve septische shock en dat de mate fibrinolytische activiteit in survivors en
nonsurvivors van een septische shock vergelijkbaar is.
ln hoofdstuk 5 wordt ingegaan op het probleem van heparine toediening bij
aortobifemorale bypass chirurgie. Tijdens deze ingrepen wordt op empirische basis
ongefractioneerd heparine (UFH) toegediend. Dit gaat gepaard met een grote variatie
in patient respons en eliminatiesnelheid, waardoor het moeilijk is voorspelbare en
adequate anticoagulantie te verkrijgen in aile patienten. In de studie wordt de
farmacokinetiek van UFH en laag moleculair heparine (LMWH) vergeleken tijdens
aonobifemorale bypass operaties. lntra-operatief vindt een significante stijging plaats
van de FbDP concentratie> in plasma. hetgeen verklaard kan worden door lysis van
intravasculaire trombi. FbDP concentraties correleren matig met de anti-Xa activiteit
in de LMWH groep, maar niet met de anti-Xa activiteit in de UFH groep. Nader
onderzoek is nodig om te bepalen. of de FbDP assay een bijdmge kan leveren bij het
152
vaststellen van een adequate LMWH dosis. Het farmacokinetisch profiel van LMWH
en UFH blijkt vergelijkbaar na een intraveneuze bolus injectie. De anti-Xa activiteit in
de LMWH groep houdt !anger aan dan die in de UFH groep.
In hoofdstuk 6 word! de waarde van de plasma FDP assays voor de diagnose van DVT
geanalyseerd op basis van gepubliceerde studies. De testeigenschappen van de D-dimer
assay voor DVT bij vermelde cutoff points zijn verschillend bij klinische patienten in
vergelijking met poliklinische patienten. Dit kan worden verklaard door een hogere
incidentie van co-morbiditeit, die interfereert met het stollingssysteem in klinische
patienten. Zeer !age D-dimer waarden zijn vrijwel bewijzend voor de afwezigheid van
DVT, terwijl zeer hoge D-dimer waarden de diagnose DVT bevestigen. Plasma FDP
assays kunnen een belangrijke rol gaan spelen in de laboratoriumdiagnose van DVT.
In de discussie wordt onderstreept, dat de nieuwe testen voor fibrine en fibrinogeen
afbraakprodukten waardevolle informatie opleveren over de heamostase in verschillende
ziektebeelden. Als directe klinische toe passing wordtde diagnostiek van DVT genoemd.
Andere mogelijke roepassingen betreffen het monitoren van het beloop van ziekten die
gepaard gaan met een gestoorde haemostase en het monitoren van therapie.
153
NAWOORD
Een proefschrift komt niet aileen tot stand door inspanningen van de schrijver, maar is
veeleer het resultaat van de inzet en medewerking van velen. Gaarne bedank ik allen
die op enigerlei wijze aan de totstandkoming ervan hebben bijgedragen.
Enkelen wil ik met name noemen.
Mijn promotor Prof. J .H.P. Wilson. ben ik dankbaar voor de mogelijkheid die hij bood
om het onderzoek op zijn afdeling uit te voeren, voor de vrijheid van onderzoek en
voor bet kritische commentaar op het manuscript.
Mijn co-promotor. Dr. Eddy Knot, verleende op geheel eigen wijze steun in de opzet
en uitvoering van het onderzoek. Eddy,jouw geestdrift en vindingrijkheid hebben enoe
bijgedragen dat ik het spoor niet bijster raakte, oak na jouw vertrek uit het ziekenhuis.
Dr. Wim van den Berg, bedankt voor de mogelijkheid die jij hood om gebruik te
kunnen maken van de faciliteiten van het lab en Dr. Felix de Rooy voor jouw
praktische adviezen op computergebied en de vele discussies. die daaruit menigmaal
voortvloeiden. Het was prettig om te weten dat ik bij problemen altijd op jullie kon
terugvallen. Samen metjullie bedank ik ook Darcos. Trinet, Annie. Rita. Marjan, Gardi,
"grate" Jan. "kleine" Jan. Miriam, Ellen. Angela en Melanie voor hun steun tijdens het
onderzoek en de plezierige tijd op het Jab.
Moniek de Maat. bedankt voor jouw hulp bij enkele studies en de laboratorium
technieken. die jij mij hebt geleerd.
Drs. Paul van Bergen. voor jouw praktische steun tijdens het onderzoek.
Dr. Willem Nieuwenhuizen dank ik voor zijn vlotte wijze van communiceren bij de
totstandkoming van artikelen en zijn constructieve adviezen daarbij.
Prof.dr. Hero van Urk en Prof.dr. Bert Eikelboom ben ik dankbaar voor hun
medewerking bij de Heparine trial.
Prof.dr.ir. Dik Habbema dank ik voor zijn kritische. constructieve commentaar op het
manuscript.
Dr. Patrick Bossuyt, voor jouw hulp bij statistische problemen en drs. Marjolein Berger,
voor jouw adviezen op besliskundig terrein.
155
De !eden van de promotiecommissie, voor hun inspanning bij de voorbereiding van de
promotie.
De stichtingen en firma's, voor hun bijdrage in de kosten van het onderzoek en/of het
drukken van bet proefschrift.
De beide paranymfen Andre Hofland en Aart Jan van der Lely dank ik voor hun morele
steun tijdens het onderzoek en hun inzet random de promotie.
Mijn ouders, die mij in staat stelden de studie Geneeskunde te voltooien en mij steeds
hebben gestimuleerd.
Margot, mijn echtgenote. voor jouw nimmer aflatende steun en geduld.
156
Curriculum vitae
De schrijver van dit proefschrift werd geboren op 27 januari 1955. De srudie
Geneeskunde werd in 1975 aangevangen aan de Erasmus Universiteit te Rotterdam,
waar hij in 1976 het propaedeutisch examen behaalde. Van 1976 tot 1983 onderbrak
hij de studie voor werkzaamheden in het bedrijfsleven. Na hervatting van de smdie in
1983 was hij als student-assistent werkzaam op de afdeling Anatomie van de Erasmus
Universiteit te Rotterdam en op de afdelingen Interne Geneeskunde I en II van het
Academisch Ziekenhuis Rotterdam-Dijkzigt. In oktober 1988 behaalde hij het
ansexamen.
Van januari 1989 tot mei 1991 was hij als wetenschappelijk onderzoeker werkzaam op
de afdeling Interne Geneeskunde ll (hoofd: Prof. J .H.P. Wilson) van het Academisch
Ziekenhuis Rotterdam-Dijkzigt en op het Centrum voor Klinische Besliskunde (hoofd:
destijds Prof.dr.J.Lubsen, later Prof.dr.ir. J.D.F. Habbema) van de Erasmus Universiteit
te Rotterdam. Aansluitend volgde hij de opleiding Interne Geneeskunde (opleider: Prof.
J.H.P. Wilson) in het Academisch Ziekenhuis Rotterdam-Dijkzigt tot november 199!.
Sedertdien is hij werkzaam als verzekeringsgeneeskundige bij de Gemeenschappelijke
Medische Dienst te Rijswijk.
159
